Tracing the History of Synanthropic Flora and Vegetation in the Czech Republic

TRACING THE HISTORY OF SYNANTHROPIC FLORA AND VEGETATION IN THE CZECH REPUBLIC

Adéla Pokorná

PhD thesis Department of Botany Charles University in Prague

2017

Supervisor: Jan Novák, PhD.

Front cover:

Silhouette of Gothic Prague, woodcut from 1493, the oldest surviving view of Prague

Lactuca serriola, photo by P. Pokorný

2 TRACING THE HISTORY OF SYNANTHROPIC FLORA AND VEGETATION IN THE CZECH REPUBLIC

Adéla Pokorná

This doctoral thesis has four enclosed papers, listed below.

Paper I

Kozáková R, Pokorný P, Mařík J, Čulíková V, Boháčová I, Pokorná A (2014) Early to high medieval colonization and alluvial landscape transformation of the Labe valley (Czech Republic): evaluation of archaeological, pollen and macrofossil evidence. Veget Hist Archaeobot 23:701–718

Paper II

Pokorná A, Houfková P, Novák J, Bešta T, Kovačiková L, Nováková K, Zavřel J, Starec P (2014) The oldest Czech fishpond discovered? An interdisciplinary approach to reconstruction of local vegetation in medieval Prague suburbs. Hydrobiologia 730:191–213

Paper III

Pokorná A, Dreslerová D, Křivánková D (2011) Archaeobotanical Database of the Czech Republic, an interim report. Interdisciplinaria Archaeologica. Natural Sciences in Archaeology 1:49–53

Paper IV

Pokorná A, Kočár P, Sádlo J, Šálková T, Žáčková P, Komárková V, Vaněček Z, Novák J (2017) Ancient and early medieval human–made habitats of the Czech Republic: Colonization history and vegetation changes. Preslia (in preparation)

Acknowledgements

This thesis would never have come into existence without all the support of my colleagues, family and friends, all of whom provided me encouragement when it was needed, which was not infrequently. Particularly, I would like to thank my supervisor Jan Novák for being always helpful and patient. I would also like to thank Jirka, Láďa, Kristýna, Radka, Petr, Jindra, Martin and many others for helpful discussion and motivation.

I also wish to thank the Czech Academy of Sciences (project No. M300020902) and especially the Institute of Archaeology of the CAS, Prague, v.v.i. for kindly supporting our work on the Archaeobotanical Database of the Czech Republic. And lastly, I must thank Steve Ridgill for correcting my English and helping me to give the text its final shape.

Declaration

I declare that I have written this thesis and that it has not been submitted for the award of a degree at any other University. All sources have been cited and clearly acknowledged.

in Prague ......

......

5 Abstract

Plant macro remains from archaeological situations were studied in order to trace the history of gradual formation of today’s synanthropic vegetation. Synanthropic plants represent a heterogeneous group of species with various qualities and strategies, as well as with various immigration histories. In general, the synanthropic flora is rich in aliens, so it is important to know, when exactly these species immigrated to our territory (to know their residence time). Besides the determination of the residence time of alien plants, also the dynamics of formation of urban flora and vegetation was studied. Special attention was paid to the Medieval Period, when the urbanisation process started. The emergence of urban agglomeration may have been the cause of the emergence of new habitats, followed by formation of new plant associations - the predecessors of the today’s ones. In general, towns represent a special case of anthropogenic environment with many various synanthropic habitats, causing their species richness. Questions 1. When exactly the synanthropic flora of medieval towns emerged? Was the transition from the Prehistory to the Medieval Period rather gradual or sudden? 2. What particular species took place in the medieval change of synanthropic vegetation? Where did these species come from? Was the emergence of medieval towns right the main cause of the observed change in species composition? 3. In which way was the local diversity of synanthropic flora influenced by medieval urbanisation? Did rather the new species emerge or, contrarily, the previously common ones extinct? 4. What factors influenced the spread of new species during the Medieval Period? Materials and Methods The data were based on the analyses of plant macroremains, sometimes in combination with other methods (mainly the pollen analysis). The case studies were focused on particular localities in Central Bohemia, tracing gradual changes taking place in each locality. To trace general trends and to answer the questions, the Archaeobotanical Database of the CR was used, covering the time span since the Neolithic to the High Middle Ages. Results and Discussion 218 archaeophytes were found in macro-remain material from high medieval towns, representing ca. 90% of unintentionally introduced alien plants. The majority of them have been present in our territory since the Prehistory; forty new aliens immigrated during the Early Medieval Period (EM). On the contrary, only several new aliens were introduced in High Medieval (HM). It means that the main wave of immigration took place in EM, which implies that the medieval immigration couldn’t be connected with the urbanisation process. Medieval urbanisation influenced plant diversity in the similar way as present processes connected with urban enlargement. The diversity of semi-natural vegetation in the suburbs drops as a result of vanishing of suitable habitats, whereas the abundance of common ruderal species increases. The increase of diversity via immigration of new alien species is connected mainly with the intensity of long-distance trade (mediated by increased propagule preasure). The frequency of alien species in medieval towns was also influenced by their residence time. The comparison of the OLD (present since the Prehistory) and NEW (since the Middle Ages) alien plants shows similar pattern as today’s comparison of archaeophytes vs. neophytes.

6 Table of Contents

1 Introduction 9 2 The aims of the thesis 13 3 Synanthropic plants, synanthropic vegetation and urban flora 14 General characteristics of synanthropic (ruderal) habitats and plants 15 Synanthropic vegetation and the adaptations of plants 16 Specifics of urban environment and diversity of urban flora 18 4 History of synathropic biotopes 20 Prehistory 20 Medieval period 22 5 Archaeobotany as a tool for studying plant diversity in the past 23 Taphonomy of macroremains in urban contexts 24 Approaches towards reconstruction of former vegetation or environment 25 Database systems 27 6 Synanthropic plants in the past 29 Residence time of alien plants 29 Immigration of aliens to Europe documented by plant macroremains 30 Records of the urban environment in archaeobotany 32 7 Summary of papers 36 Paper I 36 Paper II 39 Paper III 40 Paper IV 42 8 Discussion 44 New archaeophytes of the Medieval Period 44 Changes of plant diversity caused by medieval urbanisation 50 Does the minimum residence time (MRT) of aliens influence their frequency in medieval towns? 53 9 References 57

Paper I - Early to high medieval colonization and alluvial landscape transformation of the Labe valley (Czech Republic): evaluation of archaeological, pollen and macrofossil evidence 75

Paper II - The oldest Czech fishpond discovered? An interdisciplinary approach to reconstruction of local vegetation in medieval Prague suburbs 117

Paper III - Archaeobotanical Database of the Czech Republic, an interim report 170

Paper IV - Ancient and early medieval human–made habitats of the Czech Republic: Colonization history and vegetation changes 179

1 Introduction

The landscape transformation that occurred at the beginning of the High Middle Ages in western and central Europe was so striking and widespread that it was easily recognized in many pollen diagrams (Firbas 1949, Rösch 2000, Ralska-Jasiewiczova et al. 2004, Brown– Pluskowski 2011, Giesecke et al. 2011, Wieckowska et al. 2012, Pokorný 2011). In the Czech Republic, this radical landscape transformation occurred around the mid-13th century AD (Klápště 2012). At that time, the population was growing substantially, probably in connection with the colonization process. Large numbers of new inhabitants had created the need to establish an appropriate pattern of settlements, which has survived with only little change until today. The royal foundations of medieval towns started in the 13th century, with its heyday during the rule of Ottokar II of Bohemia. On the other hand, some of the later medieval towns probably originated somewhat earlier as proto-urban agglomerations, situated around early medieval strongholds.

We can consider the Early Middle Ages as a transitional period during which all the later transformations to come must have had their origin. Archaeological data from the Czech Republic has shown a gradual trend of concentration and extension of populated areas as early as the 6th century as a result of the increasing population, following its great decline in the Migration Period (Klápště 1994; Kuna–Profantová 2005). Despite a rather comprehensive picture of the political organization of the early medieval Czech state (Sláma 1988, Boháčová 2011) - based on early written sources (available since the beginning of the 10th century) as well as much archaeological data - our knowledge of cultural landscape development remains rather limited.

In the following text, I will concentrate on the synanthropic flora and vegetation, especially on the connection of its formation with the above outlined medieval urbanisation. The rise of urban agglomerations may have led to the establishment of new types of habitats, followed by the formation of new plant communities – the predecessors of those we know today. Taking into consideration the processes known to be connected with urbanisation today (e.g. Chocholoušková–Pyšek 2003, Pyšek et al. 2004), we may expect the following effects to have generally developed during the Middle Ages (Chytrý et al. 2009): disappearance of species of semi-natural habitats; decline of species typical for rural areas; immigration of new aliens; and increase of abundance of perennial or biennial ruderal species (mostly those associated with today’s classes Artemisietea and Galio-Urticetea).

Plant macroremains (seeds and fruits) are the main source of information we can use: since these are usually possible to be determined, according to their morphology, to the species level (in contrast to e.g. pollen grains or phytoliths). For studying synanthropic flora, the material originating from archaeological excavations from inside former settlements appears to be the most suitable (unlike that of peat-bogs, lakes and other ‘natural archives’, mostly used by quaternary palaeoecologists). Still, such material is not analogous to recent botanical data and we need to understand its limits (see, e.g. Kočár et al. 2015).

9 Firstly, archaeobotanical data are highly biased by the origin of the sediment (e.g. so-called ‘waste bias’) and by the kind of preservation of the material (e.g. the different species composition of waterlogged vs. charred assemblages)1. Secondly, the archaeobotanical sample always represents a mixture of material of highly diverse origin (with only several rare exceptions, for example: the whole sheaf containing both the harvested crops and the accompanying weeds). All together this makes the reconstruction of original plant communities quite difficult (it is, for example, often hard to distinguish between weeds and ruderals).

To cope with this, several approaches have been suggested: the Assemblage (Phytosociological) Approach (based on diagnostic species from today’s plant communities); the Individualistic (Autecological) Approach (based on Ellenberg indicator values); or the Functional Interpretation of Botanical Surveys (FIBS) - to mention those most commonly used in archaeobotany (Birks–Birks 2005, Cappers 1995, Bogaard 2004, Behre–Jacomet 1991, Schepers et al. 2013). Their advantages and complications will be discussed below, in the section Approaches towards reconstruction of former vegetation or environments. Nevertheless, several reconstructions of synanthropic vegetation of medieval towns in the Czech Republic have been undertaken by E. Opravil, using a combination of the Phytosociological and Autecological Approaches (see, e.g. Opravil 1990, 1994).

The synathropic flora is generally rich in alien plants (e.g. Chytrý et al. 2005, 2008, Williamson 1996, Kowarik 1995). It is therefore important to know the residence time of the species which have immigrated to our territory from elsewhere. However, despite the rather long tradition of archaeobotany in the Czech Republic (Čulíková 2004, Dreslerová 2008, Kočár–Dreslerová 2010), no such list has been available until now, except for Opravil (1980a- f). Unfortunately, this particular publication is in essence a popular science piece, so it provides neither the references to written sources nor any localization of the finds mentioned. Several important synthetic works dealing with the archaeobotany of Czech Prehistory focus their attention exclusively on crops and other useful plants (Tempír 1966, Kühn 1984, Wasylikowa et al. 1991, Kočár–Dreslerová 2010, Dreslerová–Kočár 2013). Obviously, these publications almost completely overlook the occurrence of wild-growing plants.

Several important synthetic works highlighting the residence times of synathropic plants are available from neighbouring countries (Willerding 1986, Rösch 1998, Lityńska–Zając 2003, Gyulai 2010). Among the synathropic plants that demonstrably immigrated to central Europe in the Medieval Period are: Amaranthus graecizans, Anthriscus caucalis, Berteroa incana, Caucalis platycarpos, Lactuca serriola, and Leonurus cardiaca. A further step towards an

1 The processes leading to the fossilisation of plant material by charring are mainly connected to crop processing after the harvest, and thus crop weeds are much more likely to be found than other plants. It has been shown elsewhere (Kočár et al. 2015), that the composition of archaeobotanical samples have been apparently biased either by the harvesting technique (mainly height of the harvest, changing during the time of harvest), or by further processing of the harvested crops (e.g. threshing or winnowing). It means that various plants have different probabilities to be found, depending on their traits (e.g. plant height or terminal velocity/size of seeds).

10 understanding of the diversity changes of anthropogenic environments (both cultivated and ruderal) from the Neolithic to Modern times has been made in eastern France (Brun 2009). The results of this analysis based on numerous palaeoenvironmental studies show that there has been a constant enrichment of the anthropogenic flora by alien species through time. Two main periods of increased immigration rate can be emphasized: the first began in the late Neolithic and reached a peak in the late Bronze Age, and the second began with the advent of Modern times.

There is an urgent need for a similar synthesis of archaeobotanical data from the Czech Republic. Many archaeobotanical analyses have already been published from both the Medieval period and Prehistory (see Chapter 6 Synanthropic plants in the past); however, a large part of the results remain unpublished. The most frequent plant species from the published analyses from medieval towns of central Europe, including the Czech Republic, are listed in table 1.

Tab. 1 The most frequent plant species of towns in central Europe, found in the archaeobotanical material from the High Middle Ages. Data based on published analyses from central Europe, including the Czech Republic. Species found in more than 50% of towns, listed alphabetically

The most frequent plant species of medieval towns in the Middle Ages Agrostemma githago, Chenopodium album agg., Euphorbia helioscopia, Fallopia convolvulus, Galeopsis bifida/tetrahit, Lapsana communis, Persicaria lapathifolia agg., Persicaria maculosa, Polygonum aviculare agg., Prunella vulgaris, Ranunculus acris, Ranunculus repens, Rumex acetosella, Rumex crispus/obtusifolius, Setaria pumila, Setaria verticillata/viridis, Solanum nigrum, Sonchus asper, Stellaria media agg., Urtica dioica, and Urtica urens.

Despite the rather extensive volume of data available, crucial questions remain unanswered. These are: (i) When exactly did the urban flora emerge? (ii) Was the change between Prehistory and the Middle Ages abrupt or gradual? (iii) What species took place in the ‘medieval change’? (iv) Did the new species emerge, or was it rather that some previously common species disappeared?

All answers to these questions are important in understanding the genesis of today’s synanthropic vegetation, as well as the connection of these processes with human activities in the past. Today, we observe many distinct changes in the synanthropic vegetation, caused both by environmental changes and the increased mobility of people and goods. It thus remains important to know whether such processes have their analogy in previous time periods. Finally, a knowledge of the residence times of alien species may help us to better understand their present-day behaviour.

Special attention should be given to the role played by the environmental conditions and human practices as regards the flora composition: in order to understand the evolution of relationships between human societies (e.g. changes in human population density and human practices), environment (e.g. changes in landscape and climate), and biodiversity through time.

11 The information provided by these analyses may improve our knowledge of the rates of change of biodiversity in the composition of past anthropogenic floras.

2 The aims of the thesis

To study the dynamics of landscape transformation that took place around the transition between the Early Middle Ages and High Middle Ages in the Czech Republic, using historical and palaeoecological data from three early medieval strongholds located in central Bohemia. These archaeological sites have radiocarbon-dated pollen and plant macrofossil evidence from oxbow sedimentary sequences which are situated in the immediate vicinity of the strongholds.

To trace the vegetation and environmental changes that took place in the medieval suburbs of Prague from the 10th to the middle of the 14th century. The permanently moist sediments of a former water reservoir that had resulted from a natural sedimentation process provided radiocarbon-dated pollen and plant macrofossils, as well as other organic material (e.g. fish and mammal bones, diatoms, algae and remains of other aquatic organisms). The data enable the reconstruction of the progressive changes from a rural site to a highly-medieval urban environment after the foundation of the New Town of Prague in 1348.

To create the Archaeobotanical Database of the Czech Republic (CZAD) as a tool for recording, archiving, disseminating and researching data on plant macroremains from archaeological sites in the Czech Republic. The process comprises: (i) Translation of the original German version of the ArboDat program. (ii) Establishment of a multilingual version, called ArboDat Multi, with English, Czech and French mutations. (iii) Collection of data - results of archaeobotanical analyses (both published and unpublished).

To reconstruct the flora of ancient human-made habitats, using the Archaeobotanical Database of the Czech Republic. Special attention will be paid to the changes of immigration rate in the time period spanning the Neolithic and Early Medieval Period. The residence times of alien plants (i.e. the time since their introduction to the territory) will be established.

To investigate possible changes in the synanthropic flora of medieval habitats, using the Archaeobotanical Database of the Czech Republic. Four time windows (10-11, 12, 13, 14-15th century) will be used to trace possible changes in the medieval synanthropic flora. Knowledge of residence times obtained from the previous analysis will be used to test their influence on species behaviour.

3 Synanthropic plants, synanthropic vegetation and urban flora

There are, in general, two main types of synanthropic habitat, both of them bearing their own flora (with, of course, some overlap): arable fields, and the ruderal habitats of human settlements and their surroundings (e.g. waste deposits, railways, verges of roads and trampled habitats). The arable field represents a very special type of habitat: heavily dependent on there being regular human activity. Its flora is well adapted to certain agricultural treatments comprised of regular ploughing and other deliberate activities taking place at specific periods of the year.

At this point I must clarify that cereal fields and their specific flora of specialised weeds will not be a focus in this thesis. Sometimes, however, it will be impossible to avoid mentioning them, partly because of the overlap of species with a wide ecological amplitude and also because of the mixed character of archaeobotanical material. Still, I should point out that the terms synanthropic and ruderal will be, to a certain extent, treated as synonymous in the following text.

Among the most typical characteristics of ruderal habitats are: an abundance of nutrients, a certain level of stress, and an amount of disturbance. Different habitats are characterised by various combinations of these three factors, their intensity and frequency. The disturbance causes a blocking of natural succession. This means that without any further disturbances the vegetation would naturally change: moving gradually from its initial state, characterised by the dominance of annuals, over the next stages dominated by biennial and perennial herbs, and on to its climax stage - the forest. Of course, such processes may also occur naturally, e.g. after landslides or in the alluvial plains of rivers, which are, in fact, just the sort of places where many ruderal plants do come from.

Besides the mentioned borrowings from naturally-disturbed habitats (a process referred to as apophytisation), the synanthropic flora contains a lot of aliens, i.e. species which have immigrated to our territory from elsewhere. The gradual rise in species richness of the ruderal vegetation during prehistoric times and the Middle Ages represents the main topic of this thesis. However, first of all we should summarise all the available information concerning: the characteristics of various ruderal biotopes; adaptations of plants for these environmental conditions; and composition of the general types of ruderal vegetation in central Europe.

General characteristics of synanthropic (ruderal) habitats and plants

14 The various synanthropic habitats differ in several characteristics, the most important being: the frequency of disturbance; the intensity of stress; and soil productivity. Plants have developed several strategies to avoid the negative effects of these types of environment. These strategies could be divided, according to Grime (1977), into several types: r (ruderal), s (stress) and c (competition), including their various combinations.

So called r-strategists are those organisms able to survive in stands with a high frequency and intensity of disturbance. These species have generally a very short life cycle and high productivity of seeds. They are typically field- or garden weeds, colonizing, at the same time, the early successional stages of highly-disturbed sites with an abundance of nutrients.

Stress, by contrast, usually doesn’t include the entire elimination of the plant, although it may be connected with its frequent damage (e.g. by intensive trampling). Other types of stress don’t include plant damage, but are connected with shortage of water or nutrients. Among the typical habitats of this type are gaps in pavements or gravelled areas along communication lines (highways or railways). Such conditions are preferred by s-strategists as they are able to resist the stressful conditions, which is one of the ways to avoid competition.

In contrast to the other two basic strategies, highly-productive habitats without disturbance are usually dominated by so-called c-strategists, i.e. robust biennial or perennial plants able to outcompete other plants by their fast rate of growth. Such conditions are typically found in abandoned grounds and gardens, former refuse dumps and old rubble heaps.

We can see that there is no typical ruderal plant possessing general characteristics. The plants differ in many traits including their level of specialisation. Those species with a wide ecological amplitude are able to grow in different types of synathropic biotopes (e.g. in cereal fields or gardens, as well as in rubble and other ruderalised sites). Thus the most abundant ruderal species belong to this category of so-called generalists, e.g. Artemisia vulgaris, Capsella bursa-pastoris, Chenopodium album Lamium album, Poa annua, Taraxacum sect. Ruderalia, Tripleurospermum inodorum, and Urtica dioica (Lososová–Simonová 2008).

Still, there are some typical traits that are common among synanthropic plants, for example: high light requirements; massive seed production; modest seedling nutritional requirements; fast-growing roots; and independence from the need of mycorrhizae. Two other traits evaluated to be typical of ruderal plants are the high requirements for nutrients and a long flowering period (Lososová–Simonová 2008). Many species of recently abundant plants are able to flower and set seed for more than four months of the year. Some common species (e.g. Capsella bursa-pastoris, Poa annua, Stellaria media) are able to flower the whole year round.

Synanthropic vegetation and the adaptations of plants

According to the newly-developed quantitative phytosociological approach (Chytrý et al. 2009), the vegetation types of ruderal habitats in the Czech Republic can be placed in the following classes: Polygono arenastri-Poëtea annuae; Stellarietea mediae; Artemisietea

15 vulgaris; and Galio-Urticetea. In the following paragraphs, the general characteristics of these classes will be summarised, drawing special attention to the adaptations of these plants.

Polygono arenastri-Poëtea annuae

The class includes disturbed vegetation of trampled habitats, formed mostly by annual plants with a ruderal or stress-tolerant life strategy. Most species are short, with prostrate growth (Polygonum arenastrum, Portulaca oleracea, Spergularia rubra and Poa annua), tough stems (Lepidium ruderale, Herniaria glabra, Prunella vulgaris, Trifolium repens, Veronica serpyllifolia), creeping stolons (Agrostis stolonifera, Cynodon dactylon, Lolium perenne, Puccinellia distans), or leaf rosettes (Bellis perennis, Leontodon autumnalis, Plantago major, Taraxacum sec. Ruderalia, and Capsella bursa-pastoris). Most of them produce many seeds, which are dispersed on human or animal feet or on vehicle wheels. Trampling leads to soil compaction, especially on loamy substrata, where long periods of desiccation can alternate with short-term surface waterlogging.

This type of vegetation is relatively poor in species diversity. Today, it grows all over the world and its floristic composition is highly uniform. It mostly represents either an initial stage of succession or the result of retrogressive succession (e.g. footpath in a lawn). The stress is caused by a combination of high availability of nutrients and a long-term desiccation of the soil. In some cases the mechanical damage of aboveground organs may be so strong, that the entire plant can be destroyed. The abundance of nutrients helps, on the other hand, to regenerate the dead biomass quickly. Original biotopes hosting this kind of vegetation are probably the paths of wild animals, as well as the surroundings of fords and watering places in alluvia (Kopecký–Hejný 1992).

Stellarietea mediae

The class includes vegetation of frequently- or recently-disturbed sites with nutrient-rich soils and a predominance of annual herbs. It occurs either as weed vegetation on arable land or as ruderal vegetation in human settlements and other disturbed places. Many species typical of this vegetation are aliens, mostly the archaeophytes (for more details see section Residence time of alien plants). Most species are short, fast-growing annuals with a ruderal strategy, producing many seeds with a permanent seed bank, often repeatedly within one growing season. They are often self-pollinated or wind-pollinated. Their short life cycle enables them to finish their development and to produce mature seeds within the short period between two devastating disturbances. Geophytes able to reproduce vegetatively by regenerating from damaged organs represent another strategy; they often have fragile underground organs (e.g. Aegopodium podagraria, Cirsium arvense, Convolvulus arvensis, Elytrigia repens).

Today, this kind of vegetation is distributed in boreal and temperate zones, as well as in sub-Mediterranean and Mediterranean zones of the Euro-Siberian area. It hosts a number of alien plants (Hyoscyamus niger, Lepidium campestre, Xanthium strumarium), some of them being escaped healing plants (Leonurus cardiaca, Marrubium vulgare, Verbena officinalis). Among the natives we may mention, for example: Chenopodium album, C. polyspermum, Elytrigia repens, Galeopsis tetrahit, Persicaria lapthifolia, Rumex acetosella, Sonchus

16 arvensis, Stellaria media and Urtica dioica. The vegetation of rubbish dumps and backfills is mostly represented by robust annuals (Atriplex oblongifolia, A. patula, Chenopodium hybridum, C. opulifolium); the surroundings of stables and fowl-runs, however, are occupied by shorter species (Anthemis cotula, Atriplex prostrata ssp. latifolia, Chenopodium vulvaria, Malva neglecta, Urtica urens).

This type of vegetation represents the initial stage of succession, whereas the subsequent stages are mostly inhibited by human activity. Without repeated disturbance of either the vegetation or the soil surface (or both), the vegetation would be regularly being replaced by other vegetation of the class Artemisietea vulgaris (on drier habitats) or Galio-Urticetea (on wetter habitats). Original biotopes of this kind of vegetation are, for example: disturbed places in alluvial plains; shallow soils on rocks or disturbed places with sparse steppe vegetation; stands strongly influenced by wild animals; watering places; and cave entrances and surroundings of holes or burrows of some mammals (Willerding 1986, Opravil 1990, Ellenberg 1996).

Artemisietea vulgaris

The class includes thermophilous vegetation of sunny and dry habitats and is composed mainly of biennial and perennial species, although some annuals are also present. It occurs chiefly in the ruderal habitats of human settlements and their surroundings, but some types are also found in semi-natural habitats. This vegetation type usually follows and replaces annual communities of Stellarietea mediae during secondary succession. In comparison with the class Galio- Urticetea, the Artemisietea vulgaris contains fewer mesophilous species and more alien plants, both archaeophytes and neophytes (see the section Residence time of alien plants).

Today, this kind of vegetation grows in warm to moderate areas throughout temperate Europe. Originally it might have occupied loessic periglacial steppes of Pleistocene cold stages or river-cut cliffs (the outside bank of a meander) and gravely deposits (Dauco-Melilotion). It contains a high share of alien plants, mostly originating from warm and dry areas. Onopordion acanthii is the most xerophilous and thermophilous type, whereas, by contrast, Arction lappae grows preferentially in moist habitats with a higher amount of nutrients.

Galio-Urticetea

The class comprises nutrient-demanding, highly-productive perennial- or, in a few cases, annual vegetation types of mesic to wet habitats, dominated by broad-leaved dicotyledons (Aegopodium podagraria, Anthriscus sylvestris, Heracleum sphondylium, Lamium album, L. maculatum, Rumex obtusifolius, Urtica dioica) or grasses (Dactylis glomerata, Elytrigia repens). Sometimes, also, annuals or shortly-lived perennials with a high demand for nutrients are present (Galium aparine, Geranium robertianum, Geum urbanum).

Many stands are species-poor and have a single dominant species. Some vegetation types are natural, for example, those growing: at the fringes of mesic forests and scrub; in canopy openings; on water banks; game trails; and in places with a high density of animals. Other stands occur in anthropogenic habitats, for example: on roadsides; in waste places; unmanaged parks

17 and gardens; or along the banks of regulated waterways. The class includes a larger proportion of native species than other types of ruderal vegetation (Simonová–Lososová 2008), and some of these are shared with mesic meadows or mesic forests. It is more common in cooler and wetter areas than the other classes of ruderal vegetation. At higher altitudes it is the most common type of ruderal vegetation.

Specifics of urban environment and diversity of urban flora

Urbanised areas represent a special case of the synathropic environment. Although the majority of factors influencing the conditions of urban habitats do not substantially differ from those of natural ones, there are obviously some clear differences between them (for a comprehensive overview, see, e.g. Jenčová 2008). In the first place, it is a specific combination of various environmental factors that makes the urban environment unique (Sukopp 1998). The main specifics are listed (according to Sukopp–Werner 1983, Sukopp 1998, Sukopp 2004, Pyšek 1996) as: high intensity of disturbances; enhanced level of nutrients; changes in soil characteristics (drying, euthrophication, compaction, changes of pH); less availability of water (drop in the underground water table due to intensive drawing on its resource and limited soaking capacity of paved surfaces); and climatic changes (0.5-2°C increase in temperature, decreased wind speed because of buildings, increased precipitation because of air pollution).

Although all of these effects can be observed, to some degree, in all synanthropic habitats, urban settlements represent a special case: due both to the long lasting intensity of these effects and the size of the area over which they act. It is not only the size itself, but also the enormous concentration of people living there, which causes the necessity of importing resources from beyond its margins, as well as the consequential high accumulation of waste. The high intensity of building activity, the existence of specific structures (such as walls, ditches, and pavements), the intensive trampling, and the rapid turnover of various goods, as well as people, are among the most important consequences of the human concentration.

The urban environment can be described as a rich mosaic of various habitats, along a gradient of increasing human influence - from its margins towards the centre (Sukopp 2004, Sukopp 1998). The urban mosaic consists of various structures, habitats and microhabitats, which creates the conditions for the coexistence of species with very different ecological demands (Sukopp–Werner 1983, Sukopp 2004). This is probably one of the main reasons why urban floras are regularly so rich. It has been repeatedly proven that the environment of urbanised areas host many more species than the surrounding landscape (e.g. Pyšek 1993, Klotz 1990, Deutschwitz et al. 2003, Kühn el al 2004). Not only do the settlements host more species, but the species diversity undeniably increases with the size2 of the settlement (Pyšek 1989, Sukopp 2004, 1998). Such a trend could be potentially explained by the increasing size as being

2 Middle sized settlements may host around 500 plant species, whereas cities above one milion inhabitants have usually more than 1300 plant species in central Europe.

18 the source of an enhanced sampling effort. Though this could be so, it has been shown that even when we compare the same size of urban and countryside area, the urban area inevitably has a higher diversity (Wania et al. 2006, Lososová et al. 2011).

Besides the heterogeneity of habitats, another factor responsible for the high diversity of towns and cities is the enhanced intensity of traffic and trade, which supports the chances for new species to be introduced there (Kowarik 1990, Trepl 1990, Pyšek 1989a). This brings us straight to the point: the immigration of alien species as another source of enhanced floristic diversity within towns. It has been demonstrated that the proportion of alien plants is much higher in towns than in rural settlements (Pyšek 1998a, 1998b), where the trend is especially marked amongst neophytes (Pyšek 1989b). Urban settlements may also serve as a source area for the further expansion of alien species to the surrounding landscape (Pyšek 1998a). At the same time, some new alien species with a decreased ability to spread spontaneously between settlements (dispersal limitation), for example, so-called casual aliens (Klotz 1990), may be unable to naturalise outside the limits of towns, which could be another factor enhancing the biodiversity of towns.

The most common plant families typically encountered in the urban settlements of temperate Europe are: Asteraceae, Poaceae, as well as Brassicaceae, Polygonaceae, Lamiaceae and Chenopodiaceae. Among the most common native plant species of European cities are: Poa annua, Polygonum aviculeare agg., and Taraxacum sect. Ruderalia. Among the most common archeophytes are: Capsella bursa-pastoris, Plantago major, and Sonchus oleraceus (Lososová et al. 2011). Notably, five of them (with the exception of S. oleraceus) are typical plants of trampled vegetation. The species having their centre of occurrence within urban areas are sometimes called urbanophilous (Sukopp–Wittig 1998), e.g. Hordeum murinum. However, the majority of frequent urban plants should be rather called urbano-neutral. This just means that they are quite common in cities; however, at the same time, they are also common in the surrounding landscape. It has been demonstrated that the urban flora of central England is mainly a mixture of common plants in the wider countryside, together with species that occur generally on waysides and in waste places (Hill et al. 2002).

4 History of synathropic biotopes

In this thesis it is the biotopes of medieval towns and their synanthropic flora that remains under the spotlight. However, in order to attempt to trace the very beginning of these phenomena, we have to go through the gradual changes in both the character of human activities and the intensity of their influence on the environment. In the following paragraphs, we will pass through the successive periods since the Palaeolithic up to the Middle Ages in order to trace the principal phenomena concerning the influence of humans on their surrounding environment, particularly in regard to what may be relevant for synanthropic vegetation.

Prehistory

We suppose, logically, that, in the remote phases of Prehistory (e.g. the Palaeolithic and perhaps also the Mesolithic, i.e. the periods without permanent settlements, at least in central Europe), human influence on the environment was only minimal. Or more precisely, that it didn’t go outside of the overlying natural processes, such as trampling on regularly visited sites or pathways, or to a certain extent some damage of trees; something not much different from the effect of big ruminants, as well as a sort of enrichment of some spots with nitrogen, similar to that of the effect we can observe in the gathering places of wild animals, or in the surroundings of their burrows.

To conclude the supposed types of human influence in the pre-Neolithic periods - we may expect a combination of several factors: disturbance; nutrient enrichment; and a certain opening of the forest canopy. And, of course, the longer a settlement lasted and the more numerous the inhabitants, the stronger the influence must have been. In the case of huntergatherers living a nomadic way of life and moving in small troops, we don’t expect there to have been any strong influence on vegetation.

Nevertheless, there are pieces of evidence (Vencl–Fridrich 2007, Svoboda 2016) suggesting some relatively extensive and apparently long-lasting settlements of mammoth- and reindeer-hunters already in the Upper Palaeolithic in Moravia (Gravettian, 30 to 20 kY BP). These people used to assemble at the crossing points of their beasts’ regular migration. It is impossible to consider, according to the available archaeological evidence, whether or not these people assembled there only temporarily and otherwise lived in smaller mobile groups during the rest of the year, although it is highly probable. Similarly, there is some evidence of Mesolithic settlements, for example, in suitable abris (cave shelters), but still we may expect only a minimal influence on the surrounding nature. We can’t exclude the fact that Mesolithic people did selectively cut some trees, which could have led to a certain change in the tree composition of their settlement surroundings, for example, the increased frequency of the hazelnut tree.

20 The introduction of the Neolithic way of life to central Europe (between 8 to 7 kY BP) implied a substantial change in the human-influenced environment. The practice of agriculture caused the emergence of a completely new biotope - the field. At the same time, the growing of crops demanded permanent settlements: to care for fields in the various seasons of the year and to store the harvest. The construction of wooden houses (sometimes just sunken ones) as well as the digging of various kinds of pits (e.g. for extraction of pot clay) caused disturbances to the soil. The increased concentration of inhabitants, as well as the centralised processing of agricultural products, caused the accumulation of waste and an increased abundance of nutrients in soils (waste was often placed in various pits dug originally for other purposes). And, of course, the more people and domestic animals, the more intensive was the trampling.

Generally, the central European settlements of the Late Stone Age had a rural character and were relatively small (Pavlů–Zápotocká 2007). A typical Neolithic settlement was usually formed of one to four wooden long houses; only rarely were there more than ten. The common number of inhabitants in one settlement during both the Neolithic and Eneolithic Periods, estimated according to finds of burials, was about two to six families (Neustupný 2008). Several finds of fortified strongholds of larger extent (up to several hectares) from the Eneolithic period were the exception rather than the rule.

From the Bronze Age onwards various changes took place. The range of crops increased (Dreslerová–Kočár 2013, Dreslerová et al 2017) and the agricultural techniques improved due to the bronze tools (Jiráň et al. 2008). We would suppose some degree of intensification of human influence due to the increasing area inhabited and also probably to a certain level of social differentiation. Settlements situated centrally at an elevated position, with their area about two to five hectares on average (in some rare cases up to several tens of hectares), then emerged in the Bronze Age (Jiráň 2008). The human population then increased, leading to higher population densities. Particularly in this period came the emergence of relatively big settlements, with their areas reaching up to tens of hectares, along with an amount of buildings (Jiráň 2008). Such exceptional settlements could probably be called proto-urban (from the point of view of the synanthropic environment).

It is important to mention the long-distance trade that was taking place in the Czech Republic from the Bronze Age onwards, which might have some share in the unintentional immigration of new plant species. This has been proven by the finds of various objects, mostly copper bars, which testify to the direct dependence of the production of bronze on international trade. The most common provenance of these finds indicates that the direction of the main influence came from the south-east. This trend also continued in subsequent periods.

Finds of iron scythes from the Iron Age (Venclová et al. 2008a) testify to the formation of a new anthropogenic biotope - hayfields (or meadows). The character of rural settlements remained similar to that of the previous period, only that the concentration of settlements became even higher (Venclová et al. 2008a,b). Relatively large (up to 100 ha) agglomerations (strongholds) with many inhabitants emerged in this period. Among these was the stronghold Vladař, one of the most thoroughly investigated by various methods of archaeobotany and palaeoecology (Pokorný et al. 2006). Intensive international contacts also continued during this

21 period (Venclová et al. 2008a,b). Some of these obviously reached as far as the Mediterranean, where, at the time, flourished ancient civilisations with large urban aglomerations (see below).

Many changes took place in Europe during the Roman epoch. The territories included within the breadth of the Roman Empire, at the time of its largest extent, were part of a large civilization centred around the Mediterranean. Within these territories emerged military garrisons, the foundations of future towns. Furthermore, agriculture was much influenced by roman habits, which may have contributed to the increased immigration of alien plants to some parts of Europe during this period (Poschlod 2015, Willerding 1986). By contrast, territories outside the Limes (limits of the Roman Empire), which is the case for the Czech Republic, experienced some kind of decline (Salač 2008). The decline of populations also continued into the subsequent Migration Period, following on after the fall of the Roman Empire. The decrease in the concentration of settlements (as well as population decline) is well documented in the Czech territory from the Migration Period (Salač 2008).

Medieval period

The arrival of the first Slavs to our territory in the middle of the 6th century AD corresponds to the beginning of the Early Medieval Period for the Czech Republic. The oldest phases remain without written sources; however, archaeologically-documented settlements show mostly a rural character with sunken huts (e.g. in Roztoky, Kuna et al. 2005). However, since the 8th Century AD, strongholds were being constructed, some of which reached a relatively large size in area. One example is the Mikulčice stronghold, with its enclosure of about 6 hectares surrounded by ramparts, and further extramural settlements surrounding the site. The place had been settled since the 8th century AD and continued until the demise of the Great Moravia Empire in the early 10th century.

Between the mid-10th and mid-12th centuries, occurred the reign of the Přemyslids in the central parts of our territory. Many strongholds were built during this particular period. Subsequently, from the 13th century onwards, the Přemyslids founded many medieval towns in the manner of the then Western Europe (see below). Later, from the 14th century on, towns grew larger and monasteries were also founded, as well as fishponds, orchards, and vineyards. During this same period, agriculture was being intensified.

5 Archaeobotany as a tool for studying plant diversity in the past

Because the results and conclusions of this thesis are largely built on data originating from the analyses of plant macrofossils, it is important to mention here some fundamental principles of the method and to point out some of its aspects which may complicate further interpretations. Along with this, some approaches that deal with these complications, in order to achieve the most objective results possible, will be mentioned below.

Archaeobotanical samples are taken during archaeological excavations. The sampling is mostly driven by a certain strategy or plan given in advance or, in some cases, taken at random (e.g. by spotting a potentially interesting context). There are various possibilities as to how sampling may be accomplished during an excavation, mostly dependent on the scientific questions being asked by the particular investigation (e.g. a small number of large-volume samples, or a large quantity of relatively small samples; sampling of all excavated contexts or just the interesting ones, etc.).

The separation technique is selected depending on the preservation condition of the macroremains. Waterlogged material is mostly separated by a wet-sieving method; dry material, by contrast, is rather separated by a flotation method. After the separated material has been dried (in some cases, however, they are preserved in a special solution to prevent them cracking during desiccation), the macroremains are sorted under a binocular microscope.

Selected diaspores are determined according to their morphology using published guides or reference collections. In most cases it is possible to determine the macroremains to species level. However, in some more complicated taxonomical groups, or in the case of the macroremains being damaged, determination only to genus level is possible (e.g. Atriplex sp.), or doublets such as Setaria verticillata/viridis are stated.

The determined macroremains are then counted - which in some cases may be slightly difficult (ambiguous). In the case of fragments, especially, these are mostly counted in the same way as if they represented intact diaspores. However, sometimes an estimation is made as to how many original diaspores resulted in the fragments found. This practice makes sense, particularly in the case of many diminutive fragments coming from relatively large diaspores (e.g. the shell of a hazelnut), because a simple summing of fragments may lead to overestimations of their true number.

Once the whole volume of a sample has been processed, it is possible to calculate the concentration of macroremains. However, this is not always the case, as often only a part of the sample volume is processed, in accord with several strategies (see, e.g. Jacomet–Kreuz 1999). The coarser fractions are mostly analyzed completely, whereas the finest fraction only in part. Various methods are used to divide the sub-samples mechanically to avoid unequal separation. In some cases, the selected macroremains - being very abundant (e.g. Papaver, Fragaria, Ficus, Rubus) - were only picked out from 10 ml of the finest fraction and their total numbers calculated afterwards (Hellwig 1997). Or the finest fraction was processed in steps of 10 ml at

23 a time, processing only continuing so long as some new taxa were recorded in the additional step.

It is clear from this short methodical overview that there are difficulties when comparing archaeobotanical data from different localities - as well as from different authors, who may differ between each other not only in their approach but also in their individual experience (i.e. their ‘personal bias’). Moreover, we have to reckon with changes in praxis over time. In the 1970s the flotation technique became a common part of archaeological excavations, whereas before that only random finds of conspicuous concentrations of macroremains, visible to the naked eye, were mostly picked. All these considerations make the study of what particular factors have influenced the plant diversity in past periods very difficult.

Taphonomy of macroremains in urban contexts

Among the other factors influencing the result of macroremains analysis belong those included in the term taphonomy. Taphonomy comprises all factors, both natural and cultural, that affect the final composition of a fossil record. Taphonomic factors include different aspects connected with: the biology of the plants; deposition of their diaspores in the sediment; the human influence on selection of species and their different treatment; the quality of macroremains’ preservation; and the re-deposition of macroremains in the soil.

Plant species may vary - in accordance with their life strategy (see above) - in their seed production as well as in their seed dispersal. These differences may afterwards influence the fossil record (e.g. r-strategists often produce high amounts of seeds, which leads to their relative overrepresentation in archaeobotanical assemblages). It is also important to note that seed and fruit production may vary in one individual species between seasons or different habitats. For example, clonal plants, in some conditions (e.g. intensive browsing), can only spread vegetatively, hence they produce no seed - even though they may be present on a site in relatively high numbers.

Depositional processes within the urban environment are very complex (Heimdahl 2005): due to the various human activities resulting in an inseparable mixture of diaspores of diverse origin (local flora, regional flora, plants from pastures and meadows, and useful plants). The local flora is considered to include plants that have grown in the immediate vicinity of the investigated area (e.g. ruderal plants and weeds). The regional flora comprises both ruderal and natural plants occurring outside the excavated area, both outside and in the town. These plants are, naturally, difficult to differentiate from seeds grown in situ, since they are often derived from similar biotopes.

Grazing by domestic animals can lead to a concentration of plant remains from pastures and meadows in urban areas, mostly by deposition of faeces and hay. However, cattle sometimes also graze ruderal plants and weeds, and, at the same time, some of the typical ruderal plants are abundant in pastures as well. Finally, many plant species that are known to have been regularly used by humans are, at the same time, abundant in habitats highly

24 influenced by human activities, e.g. Rubus sp. div, Chenopodium album, Fallopia convolvulus, or Polygonum lapathifolium.

Fossil records of crops may also differ according to which procedures have been taken in their handling. For example, cereal crops are processed through different methods of harvesting, threshing, sieving and cleaning. All these processes may result in a different selection of weed species according to plant size, diaspore size, and other plant traits, such as the terminal velocity of seeds. It is also probable that humans could have selectively affected plant communities due to their cultural habits (e.g. some species may have been cleaned away or alternatively favoured for different reasons, unknown to us).

Macrofossils are usually preserved either by waterlogging or by carbonisation (also occasionally by mineralisation). It is important to be aware of the fact that different preservation processes of the same original plant community will lead to different fossil records. Waterlogging generally leads to a wider and more complete spectra of preserved plant remains. Cereal grains and caryopses of wild grasses are more easily preserved by carbonisation, while fruits of, for example, the Apiaceae family, are much better preserved by waterlogging (Gustafsson 2000). Another example is the differences between r- and K-strategists (sensu Pianka 1970). Since r-strategists often have the capacity to form seed banks, the seeds tend to be bigger and more hardcoated than seeds from K-strategists. This may result in seeds from r- strategists being over-represented in materials with poor preservation status (Heimdahl 2005).

Samples with carbonised material are generally lacking un-carbonised material and vice versa, although there are exceptions. Our fundamental problem here is that prehistoric material from the Czech Republic is almost entirely preserved by charring, whereas the medieval material is preserved by waterlogging. It is thus logical to expect extreme complications when trying to identify a certain character and timing, as well as the extent of changes that took place exactly on the border between these two periods.

Approaches towards reconstruction of former vegetation and environment

All the taphonomical factors mentioned above make the reconstruction of original plant communities, as well as the former environment, highly difficult. Several approaches have been suggested to tackle this task and they can be divided into two groups, which have both been defined by Birks–Birks (2005): (i) the Individualistic (or Autecological) approach; and (ii) the Assemblage (or Phytosociological) approach. Both methods have been used in archaeobotany, either individually or in combination (e.g. Diekmann–Dupré 1997; Koerner et al. 1997; Persson 1981; Ter Braak–Gremmen 1987; Van der Maarel 1993, Reed 2013).

Individualistic approach

25 The Individualistic approach is based on information on environmental optima and the tolerances of a particular taxon, i.e. it examines the ecology of an individual plant species rather than the plant community as a whole. Abiotic values can be derived, for example, from Ellenberg indicator values (Ellenberg et al. 1991), based on the optimal ecological requirements of a taxon when in competition with other species. These individual, taxon-bound values may be used to reconstruct the specific abiotic conditions of an environment, such as temperature, salinity or moisture availability (Behre–Jacomet 1991; Cappers 1995). By combining different abiotic values, a taxon list can be divided into subsets that are probably sharing the same habitat.

This approach is suitable, but only so long as the response of the taxa to environmental factors does not change and that the combinations of environmental conditions between the past and nowadays are comparable (actualism), so that most probably the composition of vegetation does not, and did not, change very much over time (Schepers et al. 2013). A disadvantage of using these abiotic values is that they are based on field observations of certain growth locations, but insight into which factors are influencing the occurrence of a taxon is often lacking (Bogaard 2004, Charles et al. 1997). Therefore, Charles et al. (1997) and Bogaard (2004) proposed, in addition, the use of functional attributes (biotic factors), such as leaf life span and root length, to reconstruct those vegetation types for which a modern analogy for the combination of factors influencing the chances of a taxon occurring may be lacking: a prime example is arable weed vegetation.

Assemblage approach

The Community and Assemblage approach explores the interspecific relationships (plant sociology) of plant taxa occurring together (concurring) at a site, i.e. using the floristic composition, and not the environmental characteristics, of a habitat. The interspecific relationships of plants can be expressed in two different ways: by the ecological grouping of taxa, or by organising taxa by phytosociology. Using this approach for the interpretation of archaeobotanical material may be linked with two fundamental problems: (i) the temporal and geographical changes in the ecological co-occurrence of species are difficult to reconstruct, thus making comparisons between modern and past plant communities problematic3 (Behre– Jacomet 1991, Holzner 1978); (ii) the nature of the archaeobotanical material will be the result of various taphonomical processes (see above). Therefore, the assemblage of macroremains represents either a mere fraction of the original community, or it is potentially originating from a number of different sources.

The first way of expressing interspecific relationships is by means of the ecological grouping of taxa. Ecological taxon groups can be adopted: directly from the literature (Arnolds– Van der Maarel 1979, Ellenberg et al. 1991, Runhaar et al. 2004); by adjusting adopted taxon groups to palaeobotanical datasets (Kreuz 2005 after Ellenberg et al. 1991, Out 2012 after Arnolds–Van der Maarel 1979); or they can be constructed manually (i.e. the groups are formed

3 Some authors have even proposed past associations that are absent today, e.g. the association Bromo- Lapsanetum praehistoricum (Knörzer 1971).

26 by the individual researcher, based on his/her expert knowledge of the taxon’s current environment). Such ordering of the data into ecological groups (like ‘arable weeds’ and ‘plants of trampled places’) is particularly useful in archaeological contexts, where the relationships between human impact and ecology are an important research goal. In contrast to syntaxonomy, where concurrence is based on many actual vegetation descriptions of taxa occurring together, ecological groups have been artificially created by combining plant taxa and environmental characteristics.

The second way to organise taxa is by phytosociology. This approach aims to identify the original plant community which has resulted in the palaeobotanical dataset under study. Some authors have used diagnostic species of recent vegetation units to give evidence of the existence of various vegetation units in the past (e.g. Opravil 1990, 1994). Successful attempts to reconstruct past vegetation by modern analogues have been presented in two classic studies by Overpeck et al. (1985) and Körber-Grohne (1992). However, some authors suspect that the application of the Phytosociological approach on archaeobotanical data is inappropriate (Van der Veen 1992), as it should be based on intact plant communities (Jones 2002).

Recently, a new method of identifying past plant communities - based on a palaeobotanical dataset - has been introduced (Schepers et al. 2013), combining the present-day concurrence values of plant species to interpret macroremains data from the Neolithic. The study explores the possibility of treating a palaeobotanical sample as a sample of modern vegetation (relevé, or plot), enabling comparisons to a dataset that comprises all Dutch relevés and hence reconstruct former syntaxonomic units (plant communities). However, the method has been applied to natural vegetation and the samples analyzed didn’t originate from archaeological features but rather from palaeobotanical material.

Database systems

The increasing number of macroremains analyses in recent decades has led to the urgent need for a uniform treatment of data in an electronic form. A considerable body of archaeobotanical data is being generated with a great scientific potential - not only for archaeology but also for other fields of research such as plant history and biogeography (van Haaster–Brinkkenmper 1995). In view of this, it is of great importance to facilitate the recovery of archaeobotanical data that are dispersed among hundreds of reports, many of which are only known to a small group of scientists due to possibly being internal laboratory reports or appendices to archaeological publications.

A system that makes the storage and recovery of all archaeobotanical data more efficient becomes ever increasingly desirable. In addition, it has become more and more difficult to publish in journals huge amounts of raw data in the form of tables, together with the detailed archaeological and other related information essential for any interpretation. What is ideally required is that database programmes are used for this (Beug 1994, Tomlinson 1992, van Haaster–Brinkkemper 1995, Kreuz–Schäfer 2002).

27 The ultimate archaeobotanical database should comprise all the relevant information, i.e. not only a list of identified taxa with sample numbers and quantification, but it should also contain information about sample volume, mesh sizes, feature type, date of the context, site type and context, topographical information and bibliographical information of the reports that refer to a site. In this case what is needed is a relational database. The possible uses for such a universal relational database of archaeobotanical data would be wide: such as to obtain a supraregional or even regional overview for a particular period of time, or even of the distribution of a single species, to mention just the most evident.

The database program ArboDat (Kreuz–Schäffer 2002) is based on MS Access and enables the insertion and further processing of the results of macroremains analyses (including wood and charcoal). The uniform coding of plant taxa is fundamental to the subsequent exchange of data. A system created by S. Jacomet (Paulsen 1995) has been used as the basis for plant coding (PCODE). As the ArboDat program was originally designed for use in German- speaking countries, all the forms, tables and structural data were obviously written in the German language. Moreover, the whole database system had been designed according to the requirements of German users.

Therefore, some changes in the structural data (e.g. the list of archaeological cultural groups or phyto-geographical units) had to be made and the appearance of some forms adapted, before the program could be used effectively in the Czech Republic. These changes were incorporated into the structure of a multilingual version (called ArboDatMulti), designed by D. Křivánková. Thus CZAD – the Archaeobotanical Database of the Czech Republic - is a database of archaeobotanical data obtained by the analysis of plant macroremains from archaeological contexts. The database is currently administrated by the Institute of Archaeology in Prague. It is based on ArboDatMulti.

Although the effort to create the CZAD database has carried on since 2009, it is still a work in progress. A considerable amount of data has been already inserted (see http://www.arup.cas.cz/czad/?l=en) and the database is already being used for various evaluations and searching of data. Despite of the advantages of the database system, some problems remain to be solved. Some authors are worried about the sharing of their data or they find the database system far too complicated for their everyday use. It is also always necessary to clean and update the data, as well as to maintain the database software.

6 Synanthropic plants in the past

Residence time of alien plants

The residence time of alien plants has been identified as an important factor influencing the behaviour of species in plant communities, mainly in the degree of their invasiveness (see, e.g. Haider et al. 2010, Lososová et al. 2012a, Pyšek–Jarošík 2005, Sorte–Pyšek 2009, Wilson et al. 2007, and Williamson et al. 2009). Synathropic plants can be divided into three groups according to their origin: apophytes, archaeophytes and neophytes. It is not difficult to explain the differences between these groups, based on their assumed history, for example: the route and, above all, the timing of their immigration to a given territory.

Apophytes are native plants that have moved from their natural stands to synanthropic ones (the question remains, however, when did this occur). Both archaeophytes and neophytes are aliens for the given territory, i.e. ‘the species which have reached the area (either deliberately or accidentally) as a consequence of the activities of Neolithic or Post-neolithic man or of his domestic animals’ (Webb 1985). The two differ only by the timing of their immigration: archaeophytes have immigrated between the beginning of the Neolithic and the end of the Middle Ages, whereas neophytes have been introduced since the beginning of the Modern Period, i.e. after the year 1492 AD. The traditional classification of naturalized alien plants by their arrival time is reasonable, for it is based on changes in human behaviour (e.g. alien plants originated from different source areas before and after the discovery of America).

Much more difficult, however, is to determine the exact time of introduction of these species in the past. In the case of neophytes, we have much more detailed knowledge about their residence time than in the case of archaeophytes, as it is possible (at least in some cases) to follow written sources. Nevertheless, considering the length of the period between the beginning of the Neolithic and the end of the 15th Century AD (about 7,000 years), we need to study the group of archaeophytes in more detail (see figure 1). In particular, we need to specify their arrival time in our territory more precisely.

Moreover, the classification of species to either natives or aliens has been mainly based on their recent behaviour and occurrence in various biotopes (e.g. Webb 1985, Weber 1997, Pyšek 2003). European countries also differ in their list of species that they consider to be aliens (Germany: Kühn–Klotz 2003, Slovakia: Medvecká et al. 2012, Czech Republic: Pyšek et al. 2012, Greece: Arianoutsou et al. 2010, Italy: Celesti-Grapow et al. 2009, Estonia: Ööpik et al. 2008, Great Britain and Ireland: Preston et al. 2002, Croatia: Mitić et al. 2008, or Poland: e.g. Tokarska-Guzik et al. 2014). Many archaeophytes have come from the Near East, i.e. domesticated crops (Zohary et al. 2012) and specialised weeds of arable land (Willcox 2012), whereas other synanthropic plants are mostly less clear regarding their origin. Archaeobotany,

29 therefore, may offer an important source of information concerning the history of the synanthropic flora.

Fig. 1 Approximate proportion of plant species in the flora of the Czech Republic, according to their immigration history. The pie chart based on Danihelka et al. (2012). Numbers of taxa are including crops, subspecies, and hybrids.

Immigration of aliens to Europe documented by plant macroremains

It is generally assumed that none of the species marked down as archaeophytes immigrated to a given territory before the beginning of the Neolithic period (which follows on from the definition of the archaeophyte). Either way, scarce data exist on the occurrence of plant species in the Mesolithic period (for the Czech Republic see, e.g. Opravil 2003; Pokorný 2003; Svoboda et al. 2007, Pokorný et al. 2010), and only a minimum of them are herbs. Only Chenopodium album, Chenopodium hybridum, Rubus idaeus, R. saxatilis and Galium sp. have been reported from the Czech territory. The most valuable sources of data regarding the Mesolithic plant species spectrum are waterlogged sediments with good quality preservation of macroremains (see e.g. Bos–Urz 2003, Bos et al. 2005, Regnell 2012, Kubiak-Martens 1999, and Bishop et al. 2013).

The spread of farming practices at the beginning of the Neolithic period in Europe resulted in the immigration of alien plants from the Near East. The so-called “Neolithic crop package” (Zohary et al. 2012), i.e. the founder crops of Neolithic agriculture, evolved in SW

30 Asia about c 10,000 cal BC as a result of the long-running process of domestication (e.g. Colledge et al. 2005, Willcox 1996). The process of crop domestication was accompanied by a parallel process of specialised-weed formation (Willcox 2012). Thus, the spread of farming practices into SE Europe (into southern Greece by c 7000 cal BC, reaching Britain and Scandinavia around 3000 years later) involved not only the transport of domestic grain crops (as well as domestic animals) into areas beyond the natural range of their wild progenitor species (Colledge et al. 2004, Coward et al. 2008), but also the unintentional spread of weed species (and probably also ruderals). The observed decrease in diversity of alien wild plants on early Neolithic sites across Europe is a pattern that mimics the trend manifested in crop species (Colledge et al. 2005). Numbers of non-crop taxa represented in different regions show a gradual reduction from SW Asia (92) across to NW Europe (31).

The rapid establishment of the Early Neolithic LBK cultural complex (Linearbandkeramik or Linear Pottery Culture) across central Europe, from Hungary to Belgium and eastern France, was associated with so-called typical LBK weeds: Bromus secalinus, Bromus sterilis, Fallopia convolvulus, Galium aparine, Galium spurium, Chenopodium album, Lapsana communis, Polygonum persicaria, and Vicia hirsuta. Four of these nine species were already present in SW Asian Pre-Pottery Neolithic assemblages (B. sterilis, C. album, F. convolvulus and P. persicaria) and could have been possibly transported westwards with selected elements of the original crop package. However, C. album, F. convolvulus and P. persicaria have also been documented in Mesolithic sites in Europe (cf. e.g. Bos–Urz 2003, Negnell 2012), as well as G. aparine and L. communis (Bos–Urz 2003), both of which have no records in Neolithic SW Asia. The other species (B. sterilis and G. spurium) were probably immigrants from SE Europe, whereas in the case of Vicia hirsuta there are no records of its occurrence other than on LBK sites (Colledge et al. 2005).

A study comparing archaeobotanical data from several LBK sites in Germany and Austria to those in Bulgaria (Kreuz et al. 2005) may help us understand the process in the gradual spread of synanthropic species in the first wave of neolithisation. The aim of the study was to investigate whether the early cultivation system brought from the eastern Mediterranean region was already adapted to European conditions in the Balkans or further West. Many species they found in Early Neolithic Bulgaria reached our territory much later during the Prehistory, some of them not even before the Middle Ages (e.g. Portulaca oleracea or Sherardia arvensis).

The gradual spread of alien plants in the course of Prehistory and subsequent periods has been well documented, for example, by Willerding (1986) for central Europe, and Brun (2009) for eastern France. The results show that there has been a constant enrichment of the anthropogenic flora by alien species through time. Two main periods can be emphasized in France (Brun 2009): the first began in the late Neolithic and reached a peak in the late Bronze Age, the second began with the advent of Modern times. The enrichment of the synanthropic flora by the immigration of aliens seems to have decreased during the Roman period. In the Mediaeval period some new species appeared, but the second massive enrichment really began after the discovery of America. This constant introduction of alien species seems to have culminated at the start of the twentieth century. Since the end of the Second World War,

31 extinctions and decreases of older alien and native species have outnumbered the arrivals of new species. Archaeophytes were over-represented on arable land (almost half of the arable weeds in this study were archaeophytes), whereas, by contrast, neophytes were twice as common in ruderal habitats.

Fig. 2 - Disposition of a section of Çatal Hüyük, a prehistoric city from 6th Millennium BC with a large number of buildings clustered together (based on Hrůza 2014) Records of the urban environment in archaeobotany

The oldest investigated proto-urban settlement is Çatal Hüyük, situated 50 km southeast of Konya in south-central Anatolia, which existed from approximately 7500 BC to 5700 BC, and flourished around 7000 BC. Its population has been estimated to be, at maximum, 10,000 people (with a reasonable estimate of between 5,000 and 7,000). The site was set up as large numbers of buildings clustered together (figure 2). A diverse wild seed flora was preserved on the site in the form of macroremains – derived, at least in part, from the intentional burning of dung as fuel (Fairbairn et al. 2002). Several dry-land seed types found in the assemblage are common weeds of winter-sown crops in Anatolia (Zohary 1973), such as: Eremopyrum spp., Beta spp., Bellevalia spp., Taeniatherum caput-medusae, Sisymbrium spp., Convolvulus spp., Adonis spp. and Vaccaria pyramidata. Other important taxa found in the locality, at least at the genus level, may be associated with the common synanthropic species known also in Europe, e.g.: Anagallis sp., Arenaria spp., Artemisia vulgaris type, Atriplex spp., Bromus spp., Buglossoides arvensis

32 type, Capsella sp., Glaucium sp., Hibiscus trionum, Lepidium type, Malva spp., Onopordon sp., and Verbena sp (Fairbairn et al. 2002).

Comparable towns of exceptional age were investigated archaeologically in ancient Mesopotamia, for example, Uruk: having an area of 500 ha in 3500-2300 BC, and hosting c 50,000 inhabitants. In the first millennium BC, there were many big cities in the Near East, such as Nineveh, Nimrud and Babylon, to mention just a few examples, all of them having an area of several hundreds of hectares, tens of thousands of inhabitants, high levels of sewerage systems, parks, gardens and large market places. Unfortunately, there are only very limited archaeobotanical data from these towns available (cf., e.g. Matthews 2010), information that would be important for answering the question of whether the synanthropic vegetation of urban environments had already emerged in the ancient Near East.

Fig. 3 Rome in the third Century AD (based on Hrůza 2014)

Knossos, on the island of Crete, is considered Europe's oldest city; in the second millennium it hosted c 80,000 inhabitants (Hrůza 2014). Other ancient towns, known from the Mediterranean are, for example, Athens, Corinth, and Miletus. Rome was definitely the absolutely largest city of the ancient world (figure 3), reaching 1,386 ha in the third Century AD, and hosting one million inhabitants (Hrůza 2014). In addition, other Roman towns such as

33 Carthage, Pompeii, and Constantinople had expanses of several hundreds of hectares. Unfortunately, archaeobotanical data from these ancient towns are either lacking, or only utilitarian plants have been investigated.

However, the Roman town Mutina (predecessor of today’s Modena, Italy) was investigated archaeobotanicaly (Rinaldi et al. 2013) and the exceptionally rich data obtained from the site represent an important insight into the ancient urban flora. At least 123 wild plant species were found there, 57 of them being archaeophytes for the Czech Republic. Some of them represent the first unique find of a species that was not then introduced to central Europe until the Medieval Period (e.g. Anthriscus caucalis, Lactuca serriola, or Leonurus cardiaca). San Marino (Mercuri et al. 2009) is another important locality from the Gothic Period in the Italian peninsula (between the end of the 5th and the first half of the 6th century AD). This locality with many ruderal species (again, some of them only known in central Europe since the Middle Ages) is another example illustrating the immense importance of studying the archaeoobotany of prehistoric cities in the Mediterranean, in order to understand the genesis of the urban flora, as well as the vegetation (e.g. Bosi et al. 2015 or Mariotti Lippi et al. 2015).

Medieval towns, by contrast, are studied rather intensively by means of archaeobotany (table 2). However, seeing the disproportionality between the data existing in the Czech Republic (see, for example, CZAD: http://www.arup.cas.cz/czad/?l=en) and those published in international journals, we may consider also the available data for other countries to be only the ‘tip of the iceberg’. The published articles mostly deal with medieval diet, domestic activities and trade, as well as changes in land use and the environment in medieval times. Recently published works (e.g. Święta-Musznicka et al. 2013, Bosi et al. 2015) seem to reflect the increasing interest of archaeobotanists towards a much deeper understanding of the environmental changes taking place in the Middle Ages.

The records of urban floras of High Medieval towns in Europe are rather rich in species, being a mixture of weeds and ruderals. The most common among them (ordered according to their descending rank of ubiquity) are: Chenopodium album, Polygonum aviculare agg., Urtica dioica, Fallopia convolvulus, Galeopsis bifida/tetrahit, Persicaria lapathifolia agg., Ranunculus repens, Stellaria media aggr., Urtica urens, Euphorbia helioscopia, Prunella vulgaris, Persicaria maculosa, Rumex acetosella, Solanum nigrum, and Sonchus asper.

34 Tab 2 Examples of published archaeobotanical data from medieval urban contexts in Europe

Country Towns and references Estonia Tartu: Kihno–Hiie 2008 Finland Helsinki: Vuorela–Lempiainan 1993 Germany Göttingen: Hellwig 1995; Kiel: Wiethold 1995; Lübeck: Lynch–Paap 1986, van Haaster 1991; Überlingen: Märkle 2005; Heidelberg: Rösch 1993 Italy Ferrara: Bandini Mazzanti 2005; Florence: Mariotti Lippi et al. 2009; Modena, Parma: Bosi et al. 2011; Pompeii: Murphy et al. 2013 Lithuania Vilnius: Stančikaite et al. 2008 Norway Oslo: Griffin 1988 Poland Elbag: Latalowa et al. 2003; Kolobrzeg: Latalowa et al. 2003; Gdansk: Latalowa et al. 2009, Badura et al. 2015; Krakow: Wasylikowa 1978; Ostrów Lednicki: Polcyn 1995; Wolin: Latalowa 1999; Wroclaw: Kosina 1995 Sweden Karlstand & Norrkoping: Heimdahl 2005 the Czech Republic Prague: Beneš et al. 2002, 2012, Pokorná et al. 2014; Brno: Opravil 1990; Most: Čulíková 1987, 1995; Olomouc: Opravil 1994 Great Britain Dublin: Geraghty 1996; London: Giorgi 1997; Worcester: Greig 1981; York: Kenward–Hall 1995, Hall–Kenward 2004 the Netherlands Leiden: Vermeeren–Gumbert 2008; Amsterdam: Paap 1984

7 Summary of Papers

Paper I

Early to high medieval colonization and alluvial landscape transformation of the Labe valley (Czech Republic): evaluation of archaeological, pollen and macrofossil evidence

In the High Middle Ages, a wave of landscape transformation which originated in Western Europe swept across the east-central part of the subcontinent. In the Czech Republic, this happened during the 13th century and it had the same environmental attributes as in the rest of Europe - a considerable increase in population, vast deforestation resulting in a rapid increase in soil erosion, irreversible changes in forest species composition and the overall formation of a cultural landscape.

In the Czech Republic, the dynamics of such a radical change are, unfortunately, still only poorly understood: this paper aims to fill in this gap. Archaeological and historical data from three early medieval strongholds located in central Bohemia (Libice nad Cidlinou, Stará Boleslav and Hradišťko) are summarized and evaluated. The first two sites represent well- known political and religious centres of the early Czech state in the 10 to 11th centuries, while the last was of secondary importance.

Stará Boleslav

The area surrounding the stronghold is one of the regions which have been continuously inhabited since the Late Neolithic period (4200–2200 cal. BC). A combination of historical and archaeological sources dates the foundation of the stronghold to around 900 AD. The fortified part of the stronghold covered approximately 5 ha plus 11 ha of a densely-populated outer bailey. According to the archaeological evidence, it seems that there was no early medieval settlement immediately prior to the 9th century, when the stronghold was founded. Later, prior to the mid 11th century, Stará Boleslav became a primarily religious centre of the Czech state; however, no later than the 12th century it started to lose its importance. The continuity in the deposition of cultural layers comes to an end during the first half of the 13th century.

Before the stronghold was founded, a wooded floodplain without any settlements dominated the river valley. The first clearly-evident colonization event is demonstrated by a distinct fall in the Pinus pollen percentage. This can probably be connected with the clearance of woodland on the colonized highest fluvial terrace and, being located outside the flood zone, was rather dry and thus suitable for pines. At the same time, the woodland clearance and settlement facilitated the growth of Artemisia. An extensive landscape change occurred only later, an event that can be dated to the 11th century, and thus this section of the pollen diagram can be better connected with the foundation of the canonry, which certainly represented a significant increase in the economic and agricultural development of the site. In the 12th

36 century, the importance of Stará Boleslav as a political, religious and representative centre had already gone. However, this decline is not reflected in the pollen and macrofossil data, basically because the landscape in the region became even more intensively used.

Libice nad Cidlinou

The stronghold of Libice nad Cidlinou (hence referred to as Libice) was founded on two island- like remnants of a fluvial terrace with a total area of 24 ha. The earliest Early Medieval settlement at the site can be associated with the arrival of the Slavonic ethnic group in the course of the 6th and 7th centuries. A distinct increase in settlement activities at the site is dated to the late 9th century. The major change in settlement intensity and structure at the end of the 9th century can be interpreted as the foundation phase of the stronghold. During the 11th century there is significantly less settlement evidence at the stronghold, both in the fortified centre and its vicinity.

Even before the foundation of the stronghold, the vicinity of this site was mostly unwooded, probably reflecting the human impact of the previous settlement phase. The foundation of the stronghold was quickly followed by a distinct vegetation change, marked especially by a decrease in the trees that comprised the major woodland (Quercus, Tilia, Carpinus, Ulmus and Corylus). Furthermore, Alnus and Salix on the floodplain were cut down. During this time period the sediment type rapidly changed from predominantly organic to fine loams, probably as a consequence of the erosion that followed on from woodland clearance in the region situated upstream.

Considering the rapid change in sedimentation, we have to assume there was also a relative increase in fluvially-transported pollen, and hence an increase in the long-distance fluvial component of the pollen spectra. In our data, it is impossible to detect the landscape change which was caused only by the activities taking place around the stronghold at Libice. The local landscape change could either have been limited, as at the stronghold of Stará Boleslav, or it could have been more extensive. The latter is more likely due to the bigger expected population at Libice, and especially because there was already a settlement here before the stronghold; thus human activity had had an earlier start at this site.

Hradišťko

This settlement is situated on the east bank of the river Labe on a sand and gravel fluvial terrace remnant, approximately 7.5 km south of Libice. The total area of the settlement covered 5.2 ha. Tentatively, the foundation of this settlement can be dated to the transition from the 10th to the 11th century, and it was probably abandoned by the 12th century. The fact that the stronghold of Hradišťko has a longer and better sedimentary record and is situated in the same region as Libice (Fig. 1) compensates for the lack of a sedimentary record from the High Middle Ages at Libice. Prior to the foundation of the stronghold, there was a combination of woodland and open cultural landscape around this site. Exploitation of the landscape was not intensive but it was rather cumulative, which is reflected by the long-term gradual reduction of woodlands, during which the taxa composition remained practically undisturbed.

37 The foundation of the stronghold is reflected in the record only by the increased numbers of macrofossils of some ruderal taxa and a reduction in macrofossils of alder and oak. The rather limited local impact of the stronghold is evident from the lack of apparent change in the pollen diagram. Radical landscape transformation can only be connected with the colonization of this area in the High Middle Ages, which is dated to the late 13th or 14th century (Fig. 9), and when an accumulation of flood loams also started in the oxbow. The very limited impact of the foundation of the small stronghold of Hradišťko shown by the pollen data can serve as a rather interesting precedent for other studies: it demonstrates that especially smaller settlements in prehistoric landscapes can be invisible when traced only by pollen data.

These results suggest that the early medieval landscape in central Bohemia was still relatively well wooded. Because of these woods, individual settlements were isolated from others that were only a few kilometres distant, even when some of these settlements were important centres. We do not know if the extent and density of wooded areas is valid for the entire old settlement zone of the modern Czech Republic, but it is certainly true for the floodplain landscapes in this study.

38 Paper II

The oldest Czech fishpond discovered? An interdisciplinary approach to reconstruction of local vegetation in mediaeval Prague suburbs

Vegetation and environmental changes taking place in the medieval suburbs of Prague from the 10th to the middle of the 14th century were investigated using a multi-proxy approach. This approach combined the results of macrofossil, pollen, diatom, antracological, archaeolo- zoological and sedimentological analyses, based on sediments of a ceased water body, probably a fishpond. In the site investigated, the field indicators (both cereals and field weeds) increased over time, whereas the proportion of broad-leaf trees and shrubs decreased. This trend in the pollen spectra could be interpreted as a manifestation of the gradual enlargement of the land under plough.

At the same time, the proportion of ruderal plants increased continually - both in the pollen and macroremains spectra. The trend in grassland indicators is much less clear; however, this type of vegetation seems to decrease slightly with time. A gradual decline in the semi- natural hygrophilous vegetation was accompanied by an opposite tendency in trampled vegetation (Polygonum aviculare, Rumex acetosella). This trend, along with the distinct increase of landscape ruderalisation, seems to indicate the gradual intensification of human activity around the pond. A similar intensification of anthropogenic influence is clearly visible in the development of the aquatic environment of the pond.

During the period between the end of the 10th century to approximately the middle of the 14th century, rather extensive changes in the plant species composition took place in the region of the Prague basin. The changes observed at the investigated site clearly correspond with the general trends in landscape management reconstructed by Kozáková et al. (2009). According to this reconstruction, a fine mosaic of habitats existed in Prague before the end of the 12th century. Subsequently, this environmental diversity decreased considerably during the following centuries. This trend was explained as being the result of the increasing ruderalization and intensification of land use.

The present-day class Chenopodietea (Fumario-Euphorbion and Panico-Setarion) could be identified according to the diagnostic species. This group of vegetation showed a general tendency to increase - which culminated in the upper half of the profile. It corresponds with the pollen curve of trampled vegetation indicators and, at the same time, this trend is opposite to the trend of semi-natural types of vegetation (dry grasslands, hygrophilous herbaceous vegetation or wet meadows). When comparing this data with other data from Prague, the macroremains of these ruderal species are nearly ubiquitous - both in the archaeological sites and in natural sediments - so this vegetation probably grew inside the town and around the villages, as well as along watercourses. In contrast, perennial ruderals were very scarce on this site. Nevertheless, their particular occurrence in the Old Prague trade centre Ungelt (Opravil 1986) indicates the different environment in the urban trade centre of that time.

39 Paper III

Archaeobotanical database of the Czech Republic

CZAD was created with the intention of ensuring continuity between earlier archaeobotanical analyses, those now in progress, and those that will be undertaken in the future (Dreslerová– Pokorná 2015). At the same time, the database system is a good resource tool useful for evaluating large volumes of data, offering the possibility to obtain connected sets of information rather than that based only on one site (e.g. assessment of the first intentional production of crops, regional and supra-regional agricultural practices, or the migration of alien plant species). Moreover, a unified database system enables and promotes cooperation on an international level. The CZAD is available on the web, with versions in both Czech and English. In addition, unpublished data may be requested from the administrator; however, permission of the author is required.

URL: http://www.arup.cas.cz/czad/ (CS); http://www.arup.cas.cz/czad/?l=en (EN).

Technically, CZAD is based on the ArboDatMulti multilingual4 database program, built on the general principles, structure and thesauri of the German program ArboDat, which was developed for the processing and evaluation of archaeobotanical data (Kreuz–Schäfer 2002). It is a MS Access application, primarily designed for data processing by individual researchers, and consists of three interrelated segments. The technical tripartition of the programme has practical advantages when it comes to combining and separating data pools, as well as when updating the programme. Under normal working conditions, of course, ArboDatMulti behaves as a coherent computer application.

ArchBotProgramm contains utility programmes that include all: definitions; calculations; join-properties and operational control forms and reports: and pre-programmed queries. Additional user-defined tables, queries and other objects can be stored here.

ArchBotStrukDat contains special terms and their definitions (glossaries) ordered thematically in tables. They are usually given by abbreviations with corresponding explanations and related data. The database’s key element, plant coding, is based on a system being currently used by many archaeobotanical departments in Europe (Paulsen 1995). Additionally, the environmental characteristics of each plant species are included in the database structure.

ArchBotDaten contains the data itself, i.e. information on archaeological sites (projects), features, samples and identified plant remains. The information is ordered hierarchically in four tables (data classes): Projects (location, details on the fieldwork); Features (details on the find contexts); Samples (description of individual soil samples); and Results. The Results table contains descriptions of the identified macroremains according to their taxa, type of

4 ArboDatMulti has four language versions (English, French, German and Czech). Although some adaptations have been made to suit local requirements, the data in all versions are fully compatible with the original ArboDat programme.

40 macroremains (seed/fruit, wood, chaff, etc.) and state of preservation (charred, mineralized or waterlogged). Thus, each record corresponds to the intersection of one sample with a taxon, type of material and type of preservation; the quantity of finds in such units is recorded. In other words, if the macroremains of the same plant species in the same sample are preserved in two different ways (e.g. both charred and water-logged), they are input as two separate entries; the same rule applies when there are finds of both the seeds and vegetative parts of the same plant taxon (figure 4).

Most of the CZAD data concern seeds/fruits (i.e. carpological data), whereas wood and charcoal finds enter the database less frequently; the database is also not designed for recording pollen data. By the end of 2014, carpological data from 370 archaeological sites had been recorded, representing altogether some 7,700 samples and more than two and a half million individual macroremains.

Fig. 4 The data model of ArboDatMulti (based on Dreslerová–Pokorná 2015)

41 Paper IV

Ancient and early medieval human-made habitats of the Czech Republic: Colonization history and vegetation changes

The period between the Early Neolithic (ca. 5500 BC) and the end of the Medieval Period (1500 AD) represents a substantial part (7,000 years) of Holocene history. Our aim is to trace the development of synanthropic flora and vegetation over this time span, using archaeobotanical data. The genesis of synanthropic flora comprises both the invasions of alien species (archaeophytes, in this particular case) and the expansions of native species from populations existing in natural habitats in the surroundings. On the basis of our results, we have attempted to infer the temporal pattern and periodisation of synanthropic plants in the territory of the Czech Republic, and above all to make the residence times or, more precisely, the Minimal Residence Times (the MRTs) of the aliens more accurate.

The studied material comprises assemblages of determined macroremains from archaeologically-dated sites of the Czech Republic, covering the interval between the early Neolithic and the Early Medieval Period. The flora was reduced to those wild herbs, of alien or native origin, in which the MRT could be defined. Our objective was to demonstrate one possible application of the archaeobotanical data: its use for inferring the development scenario of synanthropic flora in remote sections of Holocene history using the Archaeobotanical Database of the Czech Republic (CZAD; Paper III).

Most herbal archaeophytes, expected to be introduced unintentionally to the Czech Republic, were found in the studied material. From 240 archaeophytes listed in current Czech flora (Danihelka et al. 2012), which obviously have never been planted intentionally, 218 were included to this study, representing ca. 90% of them. The rate of immigration of alien species changed considerably over time. Three immigration waves were distinguished, covering periods of (i) the Neolithic, (ii) the Bronze to Iron Age, and (iii) the Early Medieval Period. The dynamism of immigration waves apparently swelled during time. Especially the species-richest Early Medieval Period, though short, represented an inception of the immediately continuing immigration flow since the High Middle Ages till the present.

Several factors may caused the detected immigration rate unevenness: (i) data bias; (ii) temporal variations of propagule pressure; and (iii) successive emergence of new types of biotopes, which enabled new species to colonise them. The observed drops in immigration rate correlate with periods of documented population decline in our territory, seen as decreases in archaeological evidence. The drop of immigration rate between the Roman to Migration Period could also be connected with the relative isolation of our territory from the main donor area for the introduction of alien species, i.e. the Mediterranean Basin. The localities placed within the Limes (limits of the Empire, e.g. south Germany) may have experienced, contrarily, an increase in propagule pressure during the Roman Period (see Poschlod 2015, Willerding 1986).

The results enabled a more detailed view into vegetation ecology of the ancient cultural landscape. We distinguished nine species-groups, which newly emerged during the four

42 developmental phases (table 3). Since the vegetation of ruderalised habitats and grasslands is considerably spatially diversified, we may expect that many various vegetation types coexisted within the same settlement, and that they, besides, changed during the time in dependence to the character of human impact. Especially the succesion of ruderal vegetation is not easily interpretable without special comparison to archeological knowledge. Sequence of Dauco- Melilotion - Onopordion - Arction communities (see Tab. 3) is a possible exclusion. This chain can be interpreted by a gradual rise of nutrients in the settlement soils since the Eneolithic to the Medieval period, which may be linked e.g. to the growing intensity of human impact or to the cumulative effect due to duration of these habitats.

Tab. 3 Phases of immigration of new species with a list of ecological groups for each phase. Numerals in brackets represent number of species immigrated in a particular phase

Phase 1 - Neolithic (5600-4300 BC) Weeds of contemporary fallow land and maize or vegetable plantations (24) Annual species of trampled and/or dunged bare soils (6) Perennial species of mezic ruderal grasslands (11) Phase 2 - Eneolithic to Middle Bronze Age (4300-1300 BC) Annual cereal weeds (23) Tall biennial and perennial herbs of dry and nitrogen-poor substrata (14) Species of nutrient-rich trampled and grazed lawns (12) Phase 3 - Late Bronze Age to Migration Period (1300 BC-580 AD) Ruderal species of sunny, unevenly disturbed substrata rich in bases and organic nutrients (13) Species of meadows, pastures and dry grasslands (21) Phase 4 - Early Medieval Period (580-1200 AD) Nitrophilous ruderal species of human-made substrata (19) Pastural species avoided by animal grazing (9) Species of wet forests and alluvial meadows (9)

8 Discussion

New archaeophytes of the Medieval Period

The study in Paper IV enabled the identification of those alien plants (namely the archaeophytes) that didn’t immigrate to our territory until the Middle Ages. I identified forty species as being new arrivals for that period. Since I am interested in any possible transformation that was taking place during the Medieval Period, I would like to know whether the composition of this species group could help me deduce what environmental changes took place in that particular period. Taking into consideration the present state of knowledge (along with a number of possibly linked factors), we have to, unfortunately, restrict or thinking to just speculations at this point in time. These speculations would need to be confirmed in the future.

From the outset, we have to admit that the finds of those forty species may be just the result of an average fluctuation, caused by the uneven character of our data. I have already mentioned that our prehistoric data are mostly preserved by charring, whereas the medieval ones are preserved through waterlogging. Therefore, we can’t exclude the possibility that these species are the rare ones that, theoretically-speaking, had already been ‘present’ before the Middle Ages, but, by chance, haven’t yet been found. We can’t exclude such a scenario; still, it is not very likely, especially if we consider the situation in surrounding countries, where the same species started to be rather common exactly at the beginning of the medieval period.

The changing character of biotopes within human settlements (or the establishment of new ones) could serve as another possible explanation for the sudden emergence of new archaeophytes. Considering the situation that such a change evidently took place during the Early Middle Ages, we can’t link the immigration of new species with the emergence of medieval towns. Contrariwise, the change manifested by the immigration of new species is dated even several centuries earlier, long before the wave of royal foundations of medieval towns in our country, dated to the 13th to 15th centuries. Although the environmental parameters of early medieval strongholds probably didn’t differ substantially from that of later towns (those founded according to German town law), the same may also be true for possible differences from their earlier predecessors, e.g. the Iron Age strongholds. We may only speculate about what structures or materials (e.g. paving or lime mortar?) were connected with early medieval strongholds (and later towns), but, by contrast, were not with earlier settlement structures. In any case, to solve this question the close cooperation with an archaeologist or a historian is necessary.

Either way, the Early Medieval period is more likely just the time when a wave of these particular species spread across Europe. The process could be probably connected with the wave of urbanisation that started in Western and, above all, Southern Europe, where many towns directly followed on from the ancient cities. Our territory became an integral part of

44 Central Europe (from the political as well as economic point of view) just in the Early Middle Ages. Thus, it is highly probable that regular trade with neighbouring countries helped to mediate a relatively rapid transfer of plant diaspores (for example, either on the muddy wheels of wagons or on the hooves of draught animals, as well as in their fur). However, the idea that our target species formed an integral part of the common urban flora of early medieval towns in Western (or Southern) Europe, needs to be verified based on a larger volume of data.

The following questions are connected with the new archaeophytes of the Medieval Period: (i) Do these particular species differ substantially from the others that have immigrated during the previous periods? (ii) Where did they come from? (iii) Which way did they reach our territory and how did they proliferate? Answering these questions may help us to get closer to understanding the processes taking place in human-made habitats during the medieval period.

Traits of the new aliens

The most frequent family among the new immigrants is Asteraceae (8), followed by Lamiaceae (5), Caryophyllaceae (4), Brassicaceae and Apiaceae (3 each). They are mostly therophytes (16), less frequently hemicryptophytes (10) (Kubát et al. 2002). Taking into consideration the Ellenberg indicator values (Ellenberg et al. 1991), the species are rather xerophilous (F mostly 4), heliophilous (L mostly 7-8) and rather thermophilous (T mostly 6-7). The species are highly variable in their nutrient demands, with their N-number mostly 4-6, but often also much higher. The flowering period is mostly 3-4 months, whereas the majority of species start to flower in May-June, or later (Kubát et al. 2002). The species also highly differ in their height. Fifteen species are tall, often higher than 0.6 m (e.g. Conium maculatum, Leonurus cardiaca, Lactuca serriola, Reseda luteola), whereas ten, on the contrary, are very low (e.g. Chenopodium vulvaria, Portulaca oleracea or Sherardia arvensis).

We can see that the new immigrants represent a rather heterogenous group (table 4). The species are either field weeds (Centaurea cyanus, Caucalis platycarpos, Ranunculus arvensis, Vaccaria hispanica) or ruderals. The ruderals can be divided into two main groups: (i) the species indicating today’s unit Arction, i.e. tall and robust weeds growing on highly- fertilized and occasionally disturbed sites (Ballota nigra, Conium maculatum, Lamium album, Leonurus cardiaca. Nepeta cataria); or (ii) the today’s unit Malvion, i.e. species preferring intensively-disturbed and extremely-fertilised sites (Hordeum murinum, Anthriscus caucalis, Chenopodium vulvaria, Xanthium strumarium). Another distinctive group is represented by tall annuals growing today on waste disposal sites, e.g. Sonchus asper, Sonchus oleraceus, Amaranthus sp., Lactuca serriola, Sisymbrium officinale. The last group is formed by xerophilous pasture weeds that are avoided by grazing animals (Anchusa officinalis, Berteroa incana, Cichorium intybus, Linaria vulgaris).

Tab. 4 List of the new alien plants, which immigrated to our territory during the Middle Ages

Minimum Residence Time: EM1 Early Middle Ages 1-3 (580-950 AD); EM2 Early Middle Ages 4 (950-1200 AD); HMA High Middle Ages (1200-1500 AD) Life history: TF annual (therophyte); HKF perennial (hemicryptophyte) Ellenberg indicator values: N soil productivity or fertility; T temperature; F soil humidity; L light availability Flowering: FlMon length of flowering period (months); Flow month of the start of flowering

Height of the plant (in metres): hghMin - minimum height; hghMax maximum height Grime (1977) strategies: c competition; cr comp.-ruderal; cs comp.-stressr; r ruderal; sr stress-ruderal

46 Tab. 5 New alien plants, which immigrated to our territory during the Middle Ages - finds of selected ones during the Prehistory in various countries of Europe

Periods: NEO Neolithic;ENE Eneolithic; BRO Bronze Age;IRA Iron Age; ROM Roman Period; MP Migration Period Countries: AUS Austria; BUL Bulgaria; CRO Croatia; DEN Denmark; ENG England; FRA France; GER Germany; ITA Italy; NET the Netherlands; NOR Norway; POR Portugal; SLO Slovenia; SPA Spain; SWI Switzerland

47 Origin of the new aliens

Comparison of our data with published results from other European countries show rather an interesting pattern. Eight of the species only emerged in central Europe in the Medieval Period (Anthriscus caucalis, Berteroa incana, Caucalis platycarpos, Chenopodium vulvaria, Lactuca serriola, Leonurus cardiaca, Marrubium vulgare, Vaccaria hispanica). Interestingly, five of these eight species were recorded in Italy during the Roman or Migration Period: Anthriscus caucalis (Mercuri et al. 2009), Caucalis platycarpos (Ciaraldi 2000), Chenopodium vulvaria (Ciaraldi 2000), Marrubium vulgare (Ciaraldi 2000), Vaccaria hispanica (Alonso 2005).

Other species have isolated finds in various countries of Europe already during Prehistory; however, they never became common until the Medieval Period. The majority of them seem to have immigrated to central Europe from the Mediterranean Basin (for a general overview, see table 5). One interesting exception is Spergula arvensis, a species that is documented from Norway already in the Neolithic (Soltvedt 2000). In any case, tracing the history of migration of the synanthropic plants through Europe deserves a separate in-depth study in its own right, using not only data published in international journals, but also by cooperating in the creation of an international database based on all data available (compare, for example, Coward et al. 2008).

The path of proliferation of new immigrants in our territory

In order to trace the path of immigration and proliferation of the new immigrants, one must compare data from our sites dated to the Early Medieval Period and find out where particular species emerged first. The sites with the highest concentration of these new immigrants may have been the sources for their further proliferation. Comparing their frequency there are rather distinct differences among the species (table 6): some new immigrants were rather frequent since the very beginning of the Early Medieval Period (Ballota nigra, Centaurea cyanus, Geranium dissectum, Lamium album/maculatum, Linaria vulgaris), whereas others were encountered only rarely. Among the potential ‘gates of immigration’, various parts of today’s Prague (25 at Malá Strana, 10 in Hradčany and 8 in Staré Město) and the Mikulčice stronghold (13) were the richest in new immigrants, followed by Libice nad Cidlinou (7).

The results are consistent with what we know about the history of these sites. The Mikulčice stronghold was a significant centre of the Great Moravian Empire founded during the 9th century on the banks of the Morava River. It was an important centre of power and trade, as well as religion. Frequent international contacts (especially with the Byzantine Empire) have been documented in this particular site (e.g. Macháček et al. 2007). Another important centre was Prague, situated near a fording place over the Vltava River and, at the same time, near an intersection of several long-distance trade roads. The fact that these two localities were the richest in new plant immigrants is in accordance with the situation as known today: The sites with the highest propagule pressure, like railway transhipment stations today (e.g. Jehlik– Dostálek), represent sites with an enhanced species pool, being the source areas for the further spread of new aliens.

48 Tab. 6 New archaeophytes, which immigrated to our territory in the Early Medieval Period: Presence in early medieval sites, based on the Archaeobotanical Database of the Czech Republic. Sites ordered according to number of found species

MALÁ MALÁ STARÉ

HRAD NOVÉ

- - - -

Taxon

PRAHA STRANA MIKULČICE PRAHA PRAHA MĚSTO LIBICE CHEB PRAHA MĚSTO NADÚSTÍ LABEM BRNO UHERSKÝ BROD CHRUDIM MOS OLOMOUC Anchusa officinalis x Anthemis austriaca x Anthriscus caucalis x Atriplex patula/prostrata x x Ballota nigra x x x x x Berteroa incana x x Caucalis platycarpos x x x Centaurea cyanus x x x x x x x x x Cichorium intybus x x Conium maculatum x x x Crepis capillaris x Euphorbia peplus x Geranium dissectum x x x x x x x x Geranium molle x Chenopodium vulvaria x Lactuca serriola x x Lamium album/maculatum x x x x x Leonurus cardiaca x x Linaria vulgaris x x x x Marrubium vulgare x Microrrhinum minus x Nepeta cataria x x x Portulaca oleracea x x x Ranunculus arvensis x Reseda luteola x x x Silene dichotoma x x Silene gallica x Sisymbrium officinale x Sonchus asper x x x Sonchus oleraceus x Spergula arvensis x x Xanthium strumarium x x x

The subsequent spread of new immigrants to other sites again depended, besides other things, on the level of propagule pressure, so that the more important the stronghold was (or the more intensive its trade contacts with central sites were) the higher was the probability of new species immigration. This is in accordance with our results, which show the relative abundance of new immigrants in Libice (e.g. Ranunculus arvensis, Spergula arvensis), whereas in Hradišťko, being a small stronghold of less importance, there hasn’t been any of them identified (Paper I). The comparison between Libice and Stará Boleslav is even more interesting. The stronghold Stará Boleslav was, unlike Hradišťko, of comparable importance to Libice. Still,

49 none of the target species have been encountered there. Does this mean that an important religious centre (i.e. Stará Boleslav) had much less input of diaspores of new aliens then a centre of less religious importance, but with probably a higher trade activity? Or is the phenomenon related to the length of the settlement’s existence? Or is it all just a result of a pure accident, i.e. the result of the average position of the sampling site?

To answer these (and similar) questions conscientiously, it is necessary to bring together a thorough knowledge of history, archaeology and botany, and to select very carefully the localities for future analyses; we may thus detect various aspects of the problem (e.g. comparison of central versus rural settlements, comparison with abroad considering the trade, as well as social contacts, of a particular period, and so on).

Changes of plant diversity caused by medieval urbanisation The process of urbanization may lead to the expansion of alien plant species, as well as the decline of native species, particularly those already becoming rare (Kühn–Klotz 2006), which is mostly the pattern we see: higher species richness in cities compared to the surrounding landscape (Haeupler 1974). The influence of urbanisation on species richness has been demonstrated repeatedly by comparing the species diversity of various European towns in two sections of time separated by a time interval of at least one century. Such comparisons of historical data with the present situation has been made, for example, for: the flora of Leipzig (1867 vs. 1989, Klotz–Gutte 1992); Zürich (1839 vs. 1998, Landolt 2000); and Pilsen (1880 vs. 1990, Chocholoušková–Pyšek 2003).

In Halle, a comparison of historical and recent flora has been done over a time span of three centuries (1687 vs. 2008, Knapp et al. 2010). Within the study period, species of natural and semi-natural stands, like bogs or nitrogen-poor habitats, became extinct more often than would be expected by chance. In contrast, species dispersed by humans, and plants preferring nitrogen-rich or warm habitats, were among those newly introduced. Land-use changes, such as the transformation from agriculture to urban land-use or the drainage of bogs, were among the main drivers of these developments.

These results are in good accordance with a study taking place in Pilsen: comparing the temporal dynamics between the city and its surroundings (Chocholoušková–Pyšek 2003). The surrounding landscape with adjacent settlements was apparently less disturbed at the beginning of the study period, thus harbouring more species from semi-natural and natural vegetation. Increasing urbanization of the landscape in the second half of the 20th century reduced the total species number in the suburbs, due to the worsening of conditions for many of these species, but their retreat was not fully compensated by introductions of new species. The flora of the city itself was, on the other hand, substantially enriched by an amount of neophytes.

We may also expect similar processes in the Medieval Period: when the foundation of new towns (similarly as they have been enlarged in more recent times) could have led to a change of species composition in the local vegetation. We assume three main processes: (i)

50 disappearance (or decline) of natural and semi-natural biotopes (e.g. reduction of dry grasslands, shore vegetation and other nutrient-poor habitats, as well as reduction of woodlands as a result of logging); (ii) increased ruderalisation (e.g. enrichment in nutrients, intensive trampling, as well as enhanced frequency and extent of disturbances); (iii) introduction of new species as a result of increased propagule pressure, due to increased movement of both people and goods, above all in central settlements. In this context we mustn’t forget that it is not the absolute number of species that is important, but rather the disappearance of some taxa, which may, or may not, be compensated by some introductions of new ones, thus resulting in vegetation that is substantially changed (in the sense of different species traits).

Changing proportion of semi-natural vs. synanthropic plant species in profile A20, Rybník

number of species of number

0 1 2 3 4 5 6 7 8 9 10 11 12 13 samples

semi-natural synanthropic

Fig. 5 Changing species composition in medieval Prague suburbs from the 10th to the 14th century (profile A20, Rybník). Decline of species from natural and semi-natural habitats was not compensated by the spread of new species in ruderal vegetation

The example of the surroundings of medieval Prague (Paper II) offered the possibility of tracing the changes in species composition over several centuries (roughly from the 10th to the 14th century). The purely rural landscape around the village of Rybník changed gradually to become a periphery of today’s Old Town of Prague (founded in the middle of the 13th century) - and about one century later it happened to find itself inside the walls of the newly marked-out New Town of Prague. A combination of pollen and macroremain analyses helped to document the distinct decline of natural and semi-natural habitats, it being compensated by the spread of ruderal vegetation (figure 5), above all, the prevalence of common annual species (e.g. Polygonum aviculare and Chenopodium album). The drop in species diversity for this site hasn’t been in any way compensated by the introduction of new rare species. Regarding the so-

51 called ‘new archaeophytes’ (see above), only Centaurea cyanus and Marrubium vulgare were encountered here.

An archaeobotanical database (Paper III) helps greatly in the comparison of data from several sites and in tracing the space-time changes of their species composition. For example, when comparing data from various parts of medieval Prague a clear link between the number of new archaeophytes and the amount of trade activity has been observed. An important trade centre was originally situated in today’s Malá Strana (‘Little Quarter’; clearly already in the 10th century). Later, probably in the 12th century, there emerged a new trade centre in today’s Staré Město (Old Town of Prague), with its customs ‘offices’ and hostelries for merchants, which was later called Ungelt. Its importance as a trade centre lasted until the middle of the 14th century, when several new, large and specialised market places were established in Nové Město (New Town of Prague).

New archaeophytes in Prague: spatial and temporal changes

number of NewArch 5

0 rs.4 rs/vs vs.2

MALÁ STRANA STARÉ MĚSTO NOVÉ MĚSTO

Fig 6 Changing number of new archaeophytes (NewArch) during the Middle Ages according to quarters of historical Prague. Based on the Archaeobotanical Database of the Czech Republic.

Dating: rs.4 reign of the Přemyslid dynasty (middle 10th to middle 12th century); rs/vs the turn of 12th/13th century; vs.2 middle 14th to 15th century

Comparing the macroremains data from the three parts of medieval Prague (i.e. Malá Strana, Staré Město and Nové Město), we can begin to see a pattern (figure 6): shifting spots of maximum diversity of new archaeophytes over time, which may be interpreted as indicating increased propagule pressure linked with the transport of goods. Especially interesting, in this respect, is a comparison of the presence of new archaeophytes in the 12th century between Ungelt and the village of Rybník (the sites being less than two km apart). These results have yet to be published; however, they indicate the high potential of an archaeobotanical database

52 (such as the CZAD) for solving various historical questions, not just the past of synanthropic vegetation.

Does the minimum residence time (MRT) of aliens influence their frequency in medieval towns? In studies based on recent botanical data, comparisons of archaeophytes with neophytes show their different behaviour - as demonstrated, for example, by their different effects of homogenization/differentiation (Lososová et al. 2012b), or on their changing occurrence in cities of various size (Pyšek 1998b). Archaeophytes are generally considered to be more conservative and more stabilised in our flora, as a result of their longer residence time. Neophytes, on the other hand, are mostly rather rare - and many of them can only remain in places with a higher propagule pressure or with a sufficiently large area of convenient habitats.

The differences between these two groups are probably not merely the result of their different traits, but also their different residence times (e.g. Pyšek–Jarošík 2005). Residence time is correlated with both the acclimatisation and invasibility of alien species (e.g. Rejmánek 2000, Pyšek et al. 2009, Wilson et al. 2007). On the other hand, regulating feedbacks, such as some form of limitation via diseases or predators, are also correlated with residence time. There is good evidence that the most dramatic development of plant invasibility takes place within the first tens, or at most hundreds, of years after their immigration. However, we can’t exclude the fact that the influence of residence time remains important even on much longer time scales. It is often not known exactly when a taxon was introduced (even in the case of neophytes), so the term ‘Minimum Residence Time’ (MRT) has been suggested: the time span since the first record of a species in a country (Rejmánek 2000, Pyšek et al. 2015). Archaeophytes, on the other hand, are mostly treated as a homogenous group, although the total time span of their potential immigration is more than ten times longer than that of neophytes.

There are 1,357 alien plants in the Czech Republic (Pyšek et al. 2012), 314 of which are archaeophytes and 1,043 neophytes. The numbers may seem to indicate that we are currently experiencing an extremely high immigration rate compared with past times. However, when we exclude all aliens expected to have been introduced intentionally (i.e. useful plants) - as well as hybrids, we obtain much lower numbers: 207 archaeophytes and 416 neophytes. In figure 7, we can see how frequently the species of these two categories are encountered in our country. Archaeophytes are divided rather evenly, about one third of them being common, one third scattered, and one third rare (with several cases of ‘vanished species’, which used to be rather common before, but disappeared with recent changes in farming). Neophytes show a completely different picture, only a small proportion of them being common or scattered. Out of the total number of 416, 161 are rare and 203 are vanished, or nearly vanished, species. We may expect that many of these rare species will vanish in the future, whereas others will become more common.

53 Occurrence of alien species in current vegetation

400

350

300

250

200

150

number of species of number 100

0 archaeophytes neophytes

common scarce rare

Fig. 7 Occurrence of alien species in current vegetation of the Czech Republic (based on Pyšek et al. 2012).

Frequency: common; scattered (including locally abundant); rare (including single locality and vanished)

I would like to evaluate the macroremains data from high medieval towns to see the behaviour of aliens, as I expect a similar pattern to their behaviour in the Middle Ages, just as we see today. However, when comparing the fossil data with today’s vegetation, we have to be aware of the lowered ability of archaeobotanical data to detect rare species. The fact can be illustrated (figure 8) by the missing archaeobotanical data of a large share of those archaeophytes that are today rare (we may expect that they were also rare in the past).

More than 120 archaeophytes have been documented in Paper IV, their minimal residence time being within the range of c 7,000 years (5,600 BC - 1,500 AD). We may imagine that in the High Medieval Period, the new archaeophytes probably behaved similarly to that of neophytes today. It would be great to elaborate on this hypothesis and to test it; however, this is beyond the scope of this thesis. Still, we can see (figure 9) that the most common ones, although there are not many, are mostly the oldest immigrants (either from the Neolithic or Eneolithic). The old immigrants are also divided rather evenly, whereas the newer ones (from the Bronze Age to the Medieval Period) are mostly rare and the common ones (e.g. Centaurea cyanus) only represent a few exceptions.

54 Occurrence of archaeophytes in current vegetation: according to their residence time

35 30

number of species of number 10 5

0 OLD NEW MISS

common scarce rare

Fig. 8 Occurrence of archaophytes in current vegetation of the Czech Republic (based on Pyšek et al. 2012) according to their residence time.

Categories: OLD archaeophytes which immigrated before the Medieval Period; NEW archaeophytes that immigrated since the Early Medieval Period; MISS archaeophytes missing in archaeobotanical data from the Czech Republic

We may ask whether the different behaviour of these species is as a result of their various residence times or of their ecological demands (since the groups of species that immigrated in a single period often also share an inclination to grow in similar types of habitats - see Paper IV). Comparing recent botanical data with those from the Medieval Period may certainly help to solve that question. In any case, the first wave of immigration, taking place in the Neolithic, was comprised of mostly archaeophytes that were already common during the whole of Prehistory, as well as in the Middle Ages. At the same time, they rank among the most common synanthropic species - even today, being mostly species with a wide ecological amplitude. This is in good accordance with the behaviour of native synanthropic species, since the most common ones (both today and in the Middle Ages), for example, Polygonum aviculare, and Chenopodium album, have been documented in the majority of central European archaeological sites since the very beginning of the Neolithic, and even since the Mesolithic.

55 Occurrence of archaeophytes in archaeobotanical data from High Middle Ages: according to their residence time

number of species of number 10

0 OLD NEW

common scarce rare

Fig. 9 Occurrence of archaeophytes in archaeobotanical data from High Middle Ages (based on the Archaeobotanical Database of the Czech Republic).

Categories: common in more than 50% of sites; scarce in 10 to 50% of sites; rare in less than 10% of sites

9 References

Anonso N (2008) Crops and agriculture during the Iron Age and late antiquity in Cerdanyola del Valle`s (Catalonia, Spain). Veget Hist Archaeobot 17:75–84

Arianoutsou M, Bazos I, Delipetrou P, Kokkoris Y (2010) The alien flora of Greece: taxonomy, life traits and habitat preferences. Biol. Inv. 12:3,525–3,549

Arnolds E, Van der Maarel E (1979) De oecologische groepen in de standaardlijst van de Nederlandse flora 1975. Gorteria 9:303–312

Badura M, Możejko B, Święta-Musznicka J, Latałowa M (2015) The comparison of archaeobotanical data and the oldest documentary records (14th–15th century) of useful plants in medieval Gdańsk, northern Poland. Veget Hist Archaeobot 24(3):441–454

Bakels C (2008) Plant remains from Sardinia, Italy, with notes on barley and grape. Veget Hist Archaeobot 11:3–8

Bandini Mazzanti M, Bosi G, Mercuri AM, Accorsi CA, Guarnieri G (2005) Plant use in a city in Northern Italy during the late Mediaeval and Renaissance periods: results of the archaeobotanical investigation of “The Mirror Pit” (14th–15th century A.D.) in Ferrara. Veget Hist Archaeobot 14:442–452

Behre K-E, Jacomet S (1991) The ecological interpretation of archaeobotanical data. In: Van Zeist W, Wasylikowa K, Behre K-E (eds) Progress in old world palaeoethnobotany. 81–108, Balkema, Rotterdam

Beneš J, Čulíková V, Kosňovská J, Frolík J, Matiášek J (2012) New plants at Prague Castle and Hradčany in the early modern period: a history of selected species. Interdiscip. Archaeol 3(1):103–114

Beneš J, Kaštovský J, Kočárová R, Kočár P, Kubečková K, Pokorný P, Starec P (2002) Archaeobotany of the Old Prague Town defence system: archaeology, macro-remains, pollen and diatom analysis. Veget Hist Archaeobot 11:107–119

Beug H-J (1994) German Database on Vegetation History established at the Institute of Palynology and Quaternary Sciences, University of Göttingen. Veget Hist Archaeobot 3:63–64

Birks HH, Birks HJB (2005) Reconstructing Holocene climates from pollen and plant macrofossils. In: Mackay A, Battarbee R, Birks J, Oldfield F (eds) Global change in the Holocene. 342–357, Hodder Arnold, London

Bishop RR, Church MJ, Rowley-Conwy PA (2013) Seeds, fruits and nuts in the Scottish Mesolithic. Proc Soc Antiq Scot 143:9–71

57 Bogaard A (2004) Neolithic farming in central Europe. An archaeobotanical study of crop husbandry practices. Routledge, London

Bogaard A (2004) Neolithic farming in central Europe. An archaeobotanical study of crop husbandry practices. Routledge, London

Boháčová I (2011) Prague, Budeč and Boleslav. The reflection of state formation in early medieval archaeological sources. In: Macháček J, Ungerman Š (eds) Frühgeschichtliche Zentralorte in Mitteleuropa. (Studien zur Archäologie Europas 14). 371–395, Habelt, Bonn

Bos JAA, Urz R (2003) Late Glacial and early Holocene environment in the middle Lahn river valley (Hessen, central-west Germany) and the local impact of early Mesolithic people-pollen and macrofossil evidence. Veget Hist Archaeobot 12:19–36

Bos JAA, van Geel B, Groenewoudt BJ, Lauwerier RCGM (2005) Early Holocene environmental change, the presence and disappearance of early Mesolithic habitation near Zutphen (The Netherlands). Veget Hist Archaeobot 15:27–43

Bosi G, Bandini Mazzanti M, Florenzano A, N’siala IM, Pederzoli A, Rinaldi R, Torri P, Mercuri AM (2011) Seeds/fruits, pollen and parasite remains as evidence of site function: piazza Garibaldi-Parma (N Italy) in Roman and Mediaeval times. J Archaeol Sci 38:1,621– 1,633

Bosi G, Mercuri AM, Mazzanti MB, Florenzano A, Montecchi MC, Torri P, ... & Rinaldi R (2015) The evolution of Roman urban environments through the archaeobotanical remains in Modena–Northern Italy. J Archaeol Sci 53:19–31

Bouby L, Billaud Y (2005) Identifying prehistoric collected wild plants: a case study from Late Bronze Age settlements in the Franch Alps (Grésine, Bourget Lake, Savoie). Economic Botany 59(3) 255–267

Bouby L, Leroy F, Carozza L (1999) Food plants from late Bronze Age lagoon sites in Languedoc, southern France: reconstruction of farming economy and environment. Veget Hist Archaeobot 8:53–69

Brombacher C (1997) Archaeobotanical investigations of Late Neolithic lakeshore settlements (Lake Biel, Switzerland). Veget Hist Archaeobot 6:167–186

Brown A, Pluskowski A (2011) Detecting the environmental impact of the Baltic Crusades on a late-medieval (13th–15th century) frontier landscape: palynological analysis from Malbork Castle and hinterland, Northern Poland. J Archaeol Sci 38:1,957–1,966

Brun C (2009) Biodiversity changes in highly anthropogenic environments (cultivated and ruderal) since the Neolithic in eastern France. The Holocene 19(6):861–871

Cappers RTJ (1995) A palaeoecological model for the interpretation of wild plant species. Veget Hist Archaeobot 4:249–257

58 Cappers RTJ (1995) A palaeoecological model for the interpretation of wild plant species. Veget Hist Archaeobot 4:249–257

Celesti-Grapow L, Alessandrini A, Arrigoni PV, Banfi E, Bernardo L, Bovio M, Brundu G, Cagiotti MR, Camarda I, Carli E, Conti F, Fascetti S, Galasso G, Gubellini L, La Valva V, Lucchese F, Marchiori S, Mazzola P, Peccenini S, Poldini L, Pretto F, Prosser F, Siniscalco C, Villani MC, Viegi L,Wilhalm T, Blasi C (2009) Inventory of the non-native flora of Italy. Plant Biosystems 143:386–430

Charles M, Jones G, Hodgson JG (1997) FIBS in archaeobotany: functional interpretation of weed floras in relation to husbandry practices. J Archaeol Sci 24:1,151–1,161

Chocholoušková Z, Pyšek P (2003) Changes in composition and structure of urban flora over 120 years: a case study of the city of Plzen. Flora 198:366–376

Chytrý M, Jarošík V, Pyšek P, Hájek O, Knollová I, Tichý L, Danihelka J (2008) Separating habitat invasibility by alien plants from the actual level of invasion. Ecology 89(6):1541–1553

Chytrý M, Láníková D, Lososová Z, Sádlo J, Otýpková Z, Kočí M, ... & Hájek M (2009) Vegetace České republiky 2. Ruderální, plevelová, skalní a suťová vegetace [Vegetation of the Czech Republic 2. Ruderal, Weed, Rock and Scree Vegetation]. Academia, Praha

Chytrý M, Pyšek P, Tichý L, Knollová I, Danihelka J (2005) Invasions by alien plants in the Czech Republic: a quantitative assessment across habitats. Preslia 77:339–354

Ciaraldi M (2000) Drug preparation in evidence? An unusual plant and bone assemblage from the Pompeian countryside, Italy. Veget Hist Archaeobot 9:91–98

Colledge S, Conolly J, Shennan S (2004) Archaeobotanical evidence for the spread of farming in the eastern Mediterranean. Current Anthropology 45 (suppl.):S35–S58

Colledge S, Conolly J, Shennan S (2005) The evolution of Neolithic farming from SW Asian origins to NW European limits. European Journal of Archaeology, 8(2):137–156

Coward F, Shennan S, Colledge S, Conolly J, Collard M (2008) The spread of Neolithic plant economies from the Near East to northwest Europe: a phylogenetic analysis. J Archaeol Sci 35:42–56

Čulíková V (1987) Synanthrope Vegetation aus der mittelalterlichen Stadt Most. In: Archäologische Rettungstätigkeit in den Braunkohlengebieten und die Problematik der siedlungs-geschichtlichen Forschung. 227–231, Most

Čulíková V (1995). Rekonstruktion der synanthropen Vegetation des mittelalterlichen Most. Památky archeologické 86:83–131

Čulíková V (2004) Archeobotanika v české archeologii na prahu 3. tisíciletí. Archeologické rozhledy 56:661–671

59 Danihelka J, Chrtek Jr J, Kaplan Z (2012) Checklist of vascular plants of the Czech Republic. Preslia 84(3):647–811

Derreumaux M (2005) How to detect fodder and litter? A case study from the Roman site “Le Marais de Dourges,” France. Veget Hist Archaeobot 14:373–385

Deutschewitz K, Lausch A, Kühn I, Klotz S (2003). Native and alien plant species richness in relation to spatial heterogeneity on a regional scale in Germany. Global Ecology and Biogeography, 12(4): 299–311

Diekmann M, Dupré C (1997) Acidification and eutrophication of deciduous forests in northwestern Germany demonstrated by indicator species analysis. Journal of Vegetation Science 8(6):855–864

Dreslerová D (2008) Better later than never: environmental archaeology in the Czech Republic. In: Beneš, J., Pokorný, P. (Eds.): Bioarchaeology in the Czech Republic. 13–38, Jihočeská univerzita v Českých Budějovicích, Přírodovědecká fakulta, Archeologický ústav AVČR, Praha, České Budějovice

Dreslerová D, Kočár P (2013) Trends in cereal cultivation in the Czech Republic from the Neolithic to the Migration period (5500 B. C. – A.D. 580). Veg. Hist. Archaeobot. 22:257–268

Dreslerová D, Kočár P, Chuman T, Pokorná A (2017) Cultivation with deliberation: cereals and their growing conditions in prehistory. Veget Hist Archaeobot (In press - accepted)

Dreslerová D, Pokorná, A (2015) Archaeobotanical Database of the Czech Republic. In: Kuna M et al (Ed.) Structuring archaeological evidence: The Archaeological Map of the Czech Republic and related systems.129–134, Institute of Archaeology of the Czech Academy of Sciences, Prague, v. v. i., Prague

Ellenberg H (1996) Vegetation mitteleuropas mit den Alpen in ökologischer, dynamischer und historischer Sicht. Ed. 5. Verlag Eugen Ulmer, Stuttgart

Ellenberg H, Weber HE, Düll R, Wirth V, Werner W, Paulissen D (1991) Zeigerwerte von Pflanzen in Mitteleuropa. Scr Geobot 18:1–248

Fairbairn A, Asouti E, Near J, Martinoli D (2002) Macro-botanical evidence for plant use at Neolithic Çatalhüyük, south-central Anatolia, Turkey. Veget Hist Archaeobot 11:41–54

Figueiral I, Bouby L, Buffat L, Petitot H, Terral JF (2010) Archaeobotany, vine growing and wine producing in Roman Southern France: the site of Gasquinoy (Béziers, Hérault). Journal of Archaeological Science 37:139–149

Figueiral I, Séjalon P (2014) Archaeological wells in southern France: Late Neolithic to Roman plant remains from Mas de Vignoles IX (Gard) and their implications for the study of settlement, economy and environment. Environmental Archaeology 19(1):23–38

Firbas F (1949) Spät- und nacheiszeitliche Waldgeschichte Mitteleuropas nördlich der Alpen. I. Allgemeine Waldgeschichte. Fischer, Jena

60 Geraghty S (1996) Viking Dublin: Botanical Evidence from Fishamble Street. Royal Irish Academy, Dublin

Giesecke T, Bennett KD, Birks HJB, Bjune AE, Bozilova E, Feurdean A, Finsinger W, Froyd C, Pokorný P, Rösch M, Seppä H, Tonkov S, Valsecchi V, Wolters S (2011) The pace of Holocene vegetation change—testing for synchronous developments. Quaternary Science Reviews 30:2,805–2,814

Giorgi J (1997) Diet in late Medieval and early modern London: the archaeobotanical evidence. Society for Medieval Archaeology Monography 15, Oxbow Monography 98:197–213

Greig J (1981) The investigation of a medieval barrel-latrine from Worcester. J Archaeol Sci 8(3):265–282

Griffin K (1988) Plant remains. De arkeologiske utgravninger i Gamlebyen, Oslo 5:15–108

Grime JP (1977) Evidence for the existence of three primary strategies in plants and its relevance to ecological and evolutionary therory. The American Naturalist 111(982): 1,169– 1,194

Gustafsson S (2000) Carbonized Cereal Grains and Weed Seeds in Prehistoric Houses - an Experimental Perspective. J Archaeol Sci 27:65–70

Gyulai F (2010) Archaeobotany in Hungary, Seed, fruit, Food and Beverage Remains in the Carpathian Basin from the Neolithic to the Late Middle ages. Archaeolingua, Budapest

Haeupler H (1974) Statistische Auswertung von Punktrasterkarten der Gefäßpflanzenflora Süd- Niedersachsens. Scripta Geobot. 8:1–141

Haider S, Alexander J, Dietz H, Trepl L, Edwards PJ, Kueffer C (2010) The role of bioclimatic origin, residence time and habitat context in shaping non-native plant distributions along an altitudinal gradient. Biological Invasions 12(12):4,003–4,018

Hall A, Kenward H (2004) Setting people in their environment: plant and animal remains from Anglo-Scandinavian York. 372–426

Heimdahl J (2005) Urbanised Nature in the Past: Site formation and environmental development in two Swedish towns, AD 1200–1800. Doctoral thesis. Department of Physical Geography and Quaternary Geology, Stockholm University. Akademitryck AB, Stockholm

Hellwig M (1995) Paradieskörner Aframomum melegueta (Roscoe) K. Schum. – Ein Gewürz aus Westafrika im frühneuzeitlichen Göttingen. In: Kroll H, Pasternak R (eds.) Res archaeobotanicae. 39–48, International Workgroup for Palaeoethnobotany. Proceedings of the nineth Symposium Kiel 1992

Hellwig M (1997) Plant remains from two cesspits (15th and 16th century) and a pond (13th century) from Göttingen, southern Lower Saxony, Germany. Veget Hist Archaeobot 6:105–116

61 Henriksen PS, Robinson D (1996) Early Iron Age agriculture: archaeobotanical evidence from an underground granary at Overbygård in northern Jutland, Denmark. Veget Hist Archaeobot 5:1–11

Hill MO, Roy DB, Thompson K (2002) Hemeroby, urbanity and ruderality: bioindicators of disturbance and human impact. Journal of Applied Ecology 39:708–720

Holzner W (1978) Weed species and weed communities. Vegetatio 38(1):13–20

Hrůza J (2014) Svět měst. Academia, Praha

Jacomet S, Kreuz A (1999) Archäobotanik: Aufgaben, Methoden und Ergebnisse vegetations- und agrargeschichtlicher Forschung. Verlag Eugen Ulmer, Stuttgart (Hohenheim)

Jenčová D (2008) Synantropní květena vesnic Českobudějovické pánve. Bakalářská diplomová práce. Jihočeská univerzita v Českých Budějovicích, Přírodovědecká fakulta, České Budějovice

Jiráň L (2008) Archeologie pravěkých Čech 5. Doba bronzová. Archeologický ústav AVČR, Praha

Johnston P (2011) Plant remains from Middle Bronze Age round houses in north Cork. A Royal Irish Academy Archaeology Research Grant project in 2011

Jones G (2002) Weed ecology as a method for the archaeobotanical recognition of crop husbandry practices. Acta Palaeobotanica 42(2):185–193

Kenward HK, Hall AR (1995) Biological Evidence from 16–22 Coppergate. In: Addyman PV (ed.) The Archaeology of York. Vol 14: The Past Environment of York. 435–797

Kihno K, Hiie S (2008) Evidence of pollen and plant macro-remains from the sediments of suburban area of medieval Tartu. Estonian Journal of Archaeology 12(1):30–50

Klápště J (1994) Paměť krajiny středověkého Mostecka [Memory of the medieval landscape in the Most region]. Státní galerie výtvarného umění, Most; Ústav archeologické památkové péče severozápadních Čech, Most; Archeologický ústav AV Č R, Praha, Most

Klápště J (2012) The Czech lands in medieval transformation. Brill, Leiden

Klotz S (1990) Species area/ species/inhabitants relations in European cities. In: Sukopp H et al. (eds.). Urban Ecology. 99–104, SPB Academic. Publ., The Hague.

Klotz S, Gutte P (1992) Biologisch-ökologische Daten zur Flora von Leipzig – ein Vergleich. Acta Acad. Sci. 1:94–97

Knapp S, Kühn I, Stolle J, Klotz S (2010) Changes in the functional composition of a Central European urban flora over three centuries. Perspectives in Plant Ecology, Evolution and Systematics 12:235–244

62 Knörzer KH (1971) Urgeschichtliche Unkräuter im Rheinland ein Beitrag zur entstehungsgeschichte der Segetalgesellschaften. Vegetatio 23(3–4):89–111

Kočár P, Dreslerová D (2010) Archeobotanické nálezy pěstovaných rostlin v pravěku České republiky. Památky archeologické 101:203–242

Kočár P, Pokorná A, Komárková V (2015) Synantropní flóra pravěkých sídlišť ve světle makrozbytkové analýzy [Anthropogenic flora of prehistoric settlements according to macro- remain analysis]. Zprávy Čes. Bot. Společ., Praha 50:301–314

Koerner W, Dupouey JL, Dambrine E, Benoît M (1997) Influence of past land use on the vegetation and soils of present day forest in the Vosges mountains, France. Journal of Ecology 85(3):351–358

Kohler-Schneider M, Caneppele A (2009) Late Neolithic agriculture in eastern Austria: archaeobotanical results from sites of the Baden and Jevišovice cultures (3600–2800 B.C.). Veget Hist Archaeobot 18(1):61–74

Kooistra LI (2008) Vegetation history and agriculture in the cover-sand area west of Breda (province of Noord-Brabant, The Netherlands). Veget Hist Archaeobot 17:113–125

Kopecký K, Hejný S (1992) Ruderální společenstva bylin České republiky. Stud. Českoslov. Akad. Věd 1:1–128

Körber-Grohne U (1992) Studies in salt marsh vegetation and their relevance to the reconstruction of prehistoric plant communities. Rev Palaeobot Palynol 73:167–180

Kosina R (1995) Botanical synopsis of medieval Wroclaw. In: Kroll H Pasternak R (eds.) Res archaeobotanicae. 101–116, International Workgroup for Palaeoethnobotany. Proceedings of the nineth Symposium Kiel 1992

Kowarik I (1990) Some responses of flora and vegetation to urbanization in Central Europe. In: Sukopp H et al. (eds.), Urban Ecology. 45–74, SPB Academic. Publ., The Hague

Kowarik I (1995) On the role of alien species in urban flora and vegetation. In Pyšek P, Prach K, Rejmánek M, Wade M (eds.), Plant invasions: general aspects and special problems. 83– 103, SPB Academic Publishers, Amsterdam

Kozáková R, Pokorný P, Havrda J, Jankovská V (2009) The potential of pollen analyses from urban deposits: multivariate statistical analysis of a data set from the medieval city of Prague, Czech Republic. Veget Hist Archaeobot 18(6):477–488

Kreuz A (2005) Landwirtschaft im Umbruch? Archäobotanische Untersuchungen zu den Jahrhunderten um Christi Geburt in Hessen und Mainfranken. Ber RGK 85:97–292

63 Kreuz A, Marinova E, Schäfer E, Wiethold J (2005) A comparison of early Neolithic crop and weed assemblages from the Linearbandkeramik and the Bulgarian Neolithic cultures: differences and similarities. Veget Hist Archaeobot 14:237–258

Kreuz A, Schäfer E (2002) A new archaeobotanical database program. Veget Hist Archaeobot 11:177–179

Kubát K, Hrouda L, Chrtek J jun., Kaplan Z, Kirschner J, Štěpánek J (2002) Klíč ke květeně České republiky. Academia, Praha

Kubiak-Martens L (1999) The plant food component of the diet at the late Mesolithic (Ertebølle) settlement at Tybrind Vig, Denmark. Veget Hist Archaeobot 8:117–127

Kühn F (1984) Vývoj polních plodin a plevelů v ČSSR od neolitu po středověk [Development of field crops and weeds in the Cyechoslovakie from the Neolithic to the Middle Ages]. Sborník prací Filosofické fakulty Brněnské univerzity E 29:179–188

Kühn I, Brandl R, Klotz S (2004) The flora of German cities is naturally species rich. Evolutionary Ecol. Research 6:749–764

Kühn I, Klotz S (2003) The alien flora of Germany - basics from a new German database. In: Child L, Plant invasions: ecological threats and management solutions. 89–100, Backhuys Publishers, Leiden

Kühn I, Klotz S (2006) Urbanization and homogenization – comparing the floras of urban and rural areas in Germany. Biological Conservation 127:292–300

Kuna M, Profantová N (eds) (2005) Počátky raného středověku v Čechách [Beginning of Early Middle Ages in Bohemia]. Archeologický ústav AV Č R, Praha

Kuna M, Profantová N, Bartuška M (2005) Počátky raného středověku v Čechách: archeologický výzkum sídelní aglomerace kultury pražského typu v Roztokách. Archeologický ústav AVČR, Praha

Landolt E (2000) Some results of a floristic inventory within the city of Zürich (1984–1988). Preslia 72:441–445

Latałowa M (1999) Palaeoecological reconstruction of the environmental conditions and economy in early medieval Wolin–against a background of the Holocene history of the landscape. Acta palaeobotanica 39(2)

Latałowa M, Badura M, Jarosińska J (2003) Archaeobotanical samples from non-specific urban contexts as a tool for reconstructing environmental conditions (examples from Elblag and Kolobrzeg, northern Poland). Veget Hist Archaeobot 12:93–104

Latałowa M, Święta-Musznicka J, Pędziszewska A (2009) Żródła paleobotaniczne do rekonstrukcji wczesnych etapów rozwoju Gdańska. Środowisko–Człowiek–Cywilizacja 2:175–85

64 Lechterbeck J, Kerig T, Kleinmann A, Sillmann M, Wick L, Rösch M (2014) How was Bell Beaker economy related to Corded Ware and Early Bronze Age lifestyles?´Archaeological, botanical and palynological evidence from the Hegau, Western Lake Constance region. Environmental Archaeology 19(2):95–113

Lippi MM, Bellini C, Secci MM, Gonnelli T (2009) Comparing seeds/fruits and pollen from a Middle Bronze Age pit in Florence (Italy). J Archaeol Sci 36(5):1,135–1,141

Lityńska–Zając M (2003) Chwasty w uprawach roslinnych w pradziejach i wczesnym sredniowieczu [Segetal weeds in prehistoric and early medieval farming]. Institute of Archaeology and Ethnology, PAS, Krakow

Lososová Z, Chytrý M, Tichý L, Danihelka J, Fajmon K, Hájek O, ... & Řehořek V (2012a) Native and alien floras in urban habitats: a comparison across 32 cities of central Europe. Global Ecology and Biogeography 21(5):545–555

Lososová Z, Chytrý M, Tichý L, Danihelka J, Fajmon K, Hájek O, Kintrová K, Láníková D Otýpková Z, Řehořek V (2012b) Biotic homogenization of Central European urban floras depends on residence time of alien species and habitat types. Biological Conservation 145:179– 184

Lososová Z, Horsák M, Chytrý M, Čejka T, Danihelka J, Fajmonl K, Hájek O, Juřičková L, Kintrová K, Láníková D, Otýpková Z, Řehořek V, Tichý L (2011) Diversity of Central European urban biota: effects of human-made habitat types on plants and land snails. Journal of Biogeography 38:1152–1163

Lososová Z, Simonová D (2008) Changes during the 20th century in species composition of synanthropic vegetation in Moravia (Czech Republic). Preslia 80:291–305

Lynch A, Paap N (1986) Botanische Untersuchungen zur Grabung an der Untertrave 97 in Lübeck: ein Beitrag zu den naturräumlichen Voraussetzungen mittelalterlicher Siedlungsgeschichte. IPP

Marinova E (2013) Landwirtschaft und Landnutzung in der Region von Ovčarovo-Gorata: archäobotanische Forschungen zum Frühneolithikum in Nordostbulgarien. In: Krauss R (ed) Ovčarovo-gorata. Eine frühneolithische Siedlung in Nordostbulgarien. Archäologie in Eurasien 29. Mainz

Mariotti Lippi M, Bellini C, Mori Secci M, Gonnelli T, Pallecchi P (2015) Archaeobotany in Florence (Italy): landscape and urban development from the late Roman to the Middle Ages. Plant Biosystems-An International Journal Dealing with all Aspects of Plant Biology, 149(1):216–227

Märkle T (2005) Nutrition, aspects of land use and environment in medieval times in southern Germany: plant macro-remain analysis from latrines (late 11th–13th century A.D.) at the town of Überlingen, Lake Constance. Veget Hist Archaeobot 14:427–441

65 Matthews W (2010) Geoarchaeology and taphonomy of plant remains and microarchaeological residues in early urban environments in the Ancient Near East. Quaternary International 214:98–113

Medvecká J, Kliment J, Májeková J, Halada L, Zaliberová M, Gojdičová E, Feráková V, Jarolímek I (2012) Inventory of the alien flora of Slovakia. Preslia 84:257–309

Mercuri AM, Accorsi CA, Bandini Mazzanti M, Bigi P, Bottazzi G, Bosi G, Marchesini M, Montecchi MC, Olmi L, Pedini D (2009) From the “Treasure of Damagnano” to the archaeobotany of a Roman and Gothic settlement in the Republic of San Marino. In: Plant and Culture: seeds of the cultural heritage of Europe. 69–91, Centro Universitario per i Beni Culturali. Ravello, Edipuglia, Bari

Mitić B, Boršić I, Dujmović I, Bogdanović S, Milović M, Cigić P, ... & Nikolić T (2008) Alien flora of Croatia: proposals for standards in terminology, criteria and related database. Natura Croatica 17(2):73–90

Murphy C, Thompson G, Fuller DQ (2013) Roman food refuse: urban archaeobotany in Pompeii, Regio VI, Insula 1. Veget Hist Archaeobot 22(5):409–419

Negnell M (2012) Plant subsistence and environment at the Mesolithic site Tågerup, southern Sweden: new insights on the “Nut Age”. Veget Hist Archaeobot 21:1–16

Neustupný E (2008) Archeologie pravěkých Čech 4. Eneolit. Archeologický ústav AVČR, Praha

Ööpik M, Kukk T, Kull K, Kull T (2008) The importance of human mediation in species establishment: analysis of the alien flora of Estonia. Boreal Env. Res. 13:53–67

Opravil E (1980a) Z historie synantropní vegetace 1 [From the history of anthropogenic vegetation 1]. Živa 28:4–5

Opravil E (1980b) Z historie synantropní vegetace 2 [From the history of anthropogenic vegetation 2]. Živa 28:53–55

Opravil E (1980c) Z historie synantropní vegetace 3 [From the history of anthropogenic vegetation 3]. Živa 28:88–90

Opravil E (1980d) Z historie synantropní vegetace 4 [From the history of anthropogenic vegetation 4]. Živa 28:130–131

Opravil E (1980e) Z historie synantropní vegetace 5 [From the history of anthropogenic vegetation 5]. Živa 28:67–168

Opravil E (1980f) Z historie synantropní vegetace 6 [From the history of anthropogenic vegetation 6]. Živa 28:206–207

Opravil E (1986) Rostlinné makrozbytky z historického jádra Prahy. Archaeologica Pragensia 7:237–271

66 Opravil E (1990) Plant macro-remains from the Dominikánská ulice at Brno (features 62a, sample 39). Praha

Opravil E (1990) Společenstva synantropní vegetace ve středověku českých zemí [Die Gesellschaften der synantropen Vegetation in den böhmischen Ländern im Mittelalter]. Zprávy Kraj. vlastiv. muz. Olomouc 263:8–19

Opravil E (1994) Synantropní vegetace ze středověku a z počátku novověku města Olomouce [Synanthrope Vegetation der Stadt Olomouc aus dem Mittelalter und aus dem Anfang der Neuzeit]. Zprávy Čes. Bot. Společ., Praha 11:15–36

Opravil E (2003) Rostlinné makrozbytky [Plant macroremains]. In: Svoboda J. (ed.), Mezolit severních Čech. Komplexní výzkum skalních převisů na Českolipsku a Děčínsku, 1978–2003 [Mesolithic of Northern Bohemia. Complex excavation of rockshelters in the Česká Lípa and Děčín areas, 1978–2003]. 38–42, Archeologický ústav AVČR, Brno

Out WA (2012) What’s in a hearth? Seeds and fruits from the Neolithic fishing and fowling camp at Bergschenhoek, The Netherlands, in a wider context. Veget Hist Archaeobot 21:201– 214

Overpeck JT, Webb T III, Prentice IC (1985) Quantitative interpretation of fossil pollen spectra. Dissimilarity coefficients and the method of modern analogs. Quat Res 23:87–108

Paap NA (1984) Palaeobotanical investigations in Amsterdam. Zeist W van, Casparie WA (eds) Plants and ancient man. 339–344, Balkema, Rotterdam

Paulsen J (1995) Namen und Synonyme mitteleuropäischer Gefässpflanzen. Schweizer Botanik CD 1995, Botanisches Institut der Universität Basel

Pavlů I, Zápotocká M (2007) Archeologie pravěkých Čech 3. Neolit. Archeologický ústav AVČR, Praha

Persson S (1981) Ecological indicator values as an aid in the interpretation of ordination diagrams. Journal of Ecology 69: 71–84

Pianka ER (1970) On r and K selection. American Naturalis 104(940):592–597

Pokorná A, Dreslerová D, Křivánková D (2011) Archaeobotanical Database of the Czech Republic, an Interim Report. Interdisciplinaria Archaeologica. Natural Sciences in Archaeology 1:49–53

Pokorná A, Kočár P, Sádlo J, Šálková T, Žáčková P, Komárková V, Vaněček Z, Novák J Ancient and early medieval human-made habitats of the Czech Republic: Colonization history and vegetation changes. Preslia (in preparation)

67 Pokorný P (2003) Rostlinné makrozbytky [Plant macroremains]. In: Svoboda J. (ed.), Mezolit severních Čech. Komplexní výzkum skalních převisů na Českolipsku a Děčínsku, 1978–2003 [Mesolithic of Northern Bohemia. Complex excavation of rockshelters in the Česká Lípa and Děčín areas, 1978–2003]. 272–273, Archeologický ústav AVČR, Brno

Pokorný P (2011) Neklidné časy. Kapitoly ze společných dějin přírody a lidí. Dokořán, Praha.

Pokorný P, Boenke N, Chytráček M, Nováková K, Sádlo J, Veselý J, ... & Jankovská V (2006) Insight into the environment of a pre-Roman Iron Age hillfort at Vladař, Czech Republic, using a multi-proxy approach. Vegetation history and Archaeobotany 15(4):419–433

Pokorný P, Šída P, Chvojka O, Žáčková P, Kuneš P, Světlík I, Veselý J (2010) Palaeoenvironmental research of the Schwarzenbergem Lake, southern Bohemia, and exploratory excavations of this key Mesolithic archaeological area. Památky archeologické 101:5–38

Polcyn M (1995) Plant remains in a early medieval pot from Ostrów Lednicki. Res Archaeobotanicae. Oetker-Voges, Kiel 249–259

Popova T, Marinova E (2000) Archaeobotanical and anthracological analisys of the Roman and Early Byzantine castle Abritus in North-Eastern Bulgaria: Some palaeoethnobotanical and environmental aspects. Archaeologia Bulgarica 4(2):49–58

Poschlod P (2015) Geschichte der Kulturlandschaft. Verlag Eugen Ulmer, Stuttgart (Hohenheim)

Preston CD, Pearman DA, Dines TD (2002) New atlas of the British and Irish flora. Oxford Univ. Press, Oxford

Pyšek P (1989) On richness of Central European urban flora. Preslia, Praha 61:329–334

Pyšek P (1993) Factors affecting the diversity of flora and vegetation in European human settlements. Vegetatio, Dordrecht, 106:89–100

Pyšek P (1996) Synantropní vegetace. Vysoká škola báňská – technická univerzita, Ostrava

Pyšek P (1998a) Alien plants in Czech village flora: an analysis of species numbers. Feddes Repert. 109:139–146

Pyšek P (1998b) Alien and native species in Central European flora: a quantitative comparison. Journal Biogeogr. 25:155–163

Pyšek P (2003) How reliable are data on alien species in Flora Europaea? Flora 198:499–507

Pyšek P, Chocholoušková Z, Pyšek A, Jarošík V, Chytrý M, Tichý L (2004) Trends in species diversity and composition of urban vegetation over three decades. Journal of Vegetation Science 15:781–788

68 Pyšek P, Danihelka J, Sádlo J, Chrtek JJ, Chytrý M, Jarošík V, Kaplan Z, Krahulec F, Moravcová L, Pergl J, Štajerová K, Tichý L (2012) Catalogue of alien plants of the Czech Republic (2nd edition): checklist update, taxonomic diversity and invasion patterns. Preslia 84(2):155–255

Pyšek P, Jarošík V (2005) Residence time determines the distribution of alien plants. In: Inderjit S, Invasive plants: ecological and agricultural aspects. 77–96, Birkhäuser, Basel

Pyšek P, Křivánek M, Jarošík V (2009) Planting intensity, residence time, and species traits determine invasion success of alien woody species. Ecology 90(10):2,734–2,744

Pyšek P, Manceur AM, Alba C, McGregor KF, Pergl J, Štajerová K, ... & Lučanová M (2015) Naturalization of central European plants in North America: species traits, habitats, propagule pressure, residence time. Ecology 96(3):762–774

Ralska-Jasiewiczova M, Latałowa M, Wasylikowa K, Tobolski K, Madeyska E, Wright HE, Turner C (eds) (2004) Late Glacial and Holocene history of vegetation in Poland based on isopollen maps. W. Szafer Institute of Botany, Polish Academy of Sciences, Kraków

Reed K (2013) Farmers in Transition: The archaeobotanical analysis of the Carpathian Basin from the Late Neolithic to the Late Bronze Age (5000–900 BC). Doctoral thesis. Department of Philosophy, University of Leicester, Leicester

Rejmánek M (2000) Invasive plants: approaches and predictions. Austral Ecol 25:497–506

Rinaldi R, Bandini Mazzanti M, Bosi G (2013) Archaeobotany in urban sites: the case of Mutina. Ann. Bot. (Roma) 3:217–230

Rösch M (1993) Pflanzenreste aus einer spätmittelalterlichen Latrine und aus einem Keller der Frühen Neuzeit im Bereich des ehemaligen Augustinerklosters in Heidelberg

Rösch M (1998) The history of crops and crop weeds in south–western Germany from the Neolithic period to modern times, as shown by archaeobotanical evidence. Veget Hist Archaeobot 7:109–125

Rösch M (2000) Long-term human impact as registered in an upland pollen profile from southern Black Forest, south-western Germany. Veget Hist Archaeobot 9:205–218

Runhaar J, Van Landuyt W, Groen CLG, Weeda EJ, Verloove F (2004) Herziening van de indeling in ecologische soortengroepen voor Nederland en Vlaanderen. Gorteria 30:12–26

Salač V (2008) Archeologie pravěkých Čech 8. Doba řimská a stěhování národů. Archeologický ústav AVČR, Praha

Schepers M, Scheepens JF, Cappers RTJ, van Tongeren OFR, Raemaekers DCM, Bekker RM (2013) An objective method based on assemblages of subfossil plant macro-remains to reconstruct past natural vegetation: a case study at Swifterbant, The Netherlands. Veget Hist Archaeobot 22:243–255

69 Schmidl A, Oeggl K (2005) Subsistence strategies of two Bronze Age hill-top settlements in the eastern Alps—Friaga/Bartholom berg (Vorarlberg, Austria) and Ganglegg/Schluderns (South Tyrol, Italy). Veget Hist Archaeobot 14:303–312

Sláma J (1988) Střední Čechy v raném středověku. 3. Archeologie o počátcích přemyslovského státu [Central Bohemia during the Early Middle Ages. Archaeology of the early Czech state], (Praehistorica 14) Univerzita Karlova, Praha

Soltvedt EC (2000) Carbonised Cereal from Three Late Neolithic and Two Early Bronze Age Sites in Western Norway. Environmental Archaeology 5:49–62

Sorte FAL, Pyšek P (2009) Extra‐regional residence time as a correlate of plant invasiveness: European archaeophytes in North America. Ecology 90(9):2,589–2,597

Šoštarić R, Alegro A, Hršak V, Stančić Z, Küster H (2009) Plant Remains from an Early Iron Age Well at Hajndl, Slovenia. Coll. Antropol. 33(4): 1295–1301

Šoštarić R, Küster H (2001) Roman plant remains from Veli Brijun (island of Brioni), Croatia. Veget Hist Archaeobot 10:227–233

Stančikaitė M, Kisielienė D, Mažeika J, Blaževičius P (2008) Environmental conditions and human interference during the 6th and 13th–15th centuries A.D. at Vilnius Lower Castle, east Lithuania. Veget Hist Archaeobot 17(1):239–250

Stika HP, Heiss AG, Zach B (2008) Plant remains from the early Iron Age in western Sicily: differences in subsistence strategies of Greek and Elymian sites. Veget Hist Archaeobot 17(1):139–148

Sukopp H (1998) Urban Ekology - Scientific and Practical Aspects. In: Breuste ., Feldmann H, Ullmann O (eds.), Urban Ekology. 1–16, Springer Verlag Berlin, Heidelberg

Sukopp H (2004) Human-caused impact on preserved vegetation. Landscape and urban planning 68(4):347–355

Sukopp H, Werner P (1983) Urban environment and vegetation. In: Holzner W et al. (ed.), Man´s impact on vegetation. 247–260, Junk, The Hague

Sukopp H, Wittig R (1998) Stadtökologie. Ein Fachbuch fur Studium und Praxis. Fischer, Stuttgart

Svoboda JA (2016) Pavlov a Doní Věstonice, Academia, Praha

Svoboda JA, Hajnalová M, Horáček I, Novák M, Přichystal A, Šajnerová A, Yaroshevich A (2007) Mesolithic settlement and activities in rockshelters of the Kamenice river canyon, Czech Republic. Eurasian Prehistory 5:95–127

Święta-Musznicka J, Latałowa M, Badura M, Gołembnik A (2013) Combined pollen and macrofossil data as a source for reconstructing mosaic patterns of the early medieval urban habitats - a case study from Gdańsk, N. Poland. J Archaeol Sci 40:637–648

70 Tempír Z (1966) Výsledky paleoetnobotanického studia pěstování zemědělských plodin na území ČSSR [Results of palaeobotanical research of cultivated crops from the territory of ČSSR]. Vědecké práce Československého zemědělského muzea 6:27–144

Ter Braak, CJF, Gremmen NJM (1987) Ecological amplitudes of plant species and the internal consistency of Ellenberg's indicator values for moisture. Vegetatio 69(1–3):79–87

Tereso JP, Ramil-Rego P, De Caravalho TP, Van Zeist W, Palfenier-Vegter RM (2015) The archaeobotany of Peelo. 3. Iron Age and Roman period. Palaeohistoria 481-490

Tokarska-Guzik B, Bzdęga K, Nowak T, Lewandowska A, Gancarek M, Frelich M (2014) Alien plants in Poland: research directions and putting the results into practice. Biodiversity Research and Conservation 35(1):57–74

Tomlinson P (1992) Design and implementation of a relational database for archaeobotanical records from Great Britain and Ireland. Circaea 10(1):1–30

Trepl L (1990) Research on the anthropogenic migration of plants and naturalization. Its history and current state of development. In: Sukopp H et al. (eds.), Urban Ecology. 75–98, SPB Academic. Publ., The Hague

Van der Maarel E (1993) Relations between sociological–ecological species groups and Ellenberg indicator values. Phytocoenologia 23:343–362

Van der Veen M (1992) Crop husbandry regimes: an archaeobotanical study of farming in northern England, 1000 BC–AD 500. JR Collis publications, Sheffield van Haaster H (1991) Umwelt und Nahrungswirtschaft in der Hansestadt Lubeck vom 12. jahrhundert bis in die Neuzeit. Lubecker Schriften zur Archaeologi und Kulturgeschichte (LSAK) 21:203–222 van Haaster H, Brinkkenmper O (1995) RADAR, a Relational Aarchaeobotanical Database for Advanced Research. Veget Hist Archaeobot 4:117–125

Vencl S, Fridrich J (2007) Archeologie pravěkých Čech 2. Paleolit a mezolit. Archeologický ústav AVČR, Praha

Venclová N (2008a) Archeologie pravěkých Čech 6. Doba halštatská. Archeologický ústav AVČR, Praha

Venclová N (2008b) Archeologie pravěkých Čech 7. Doba laténská. Archeologický ústav AVČR, Praha

Vermeeren C, Gumbert B (2008) ‘‘Ontellicke boomen’’ or ‘‘countless trees’’: reconstructing the late medieval vegetation surrounding the 16th century St Margaretha Convent, Leiden, The Netherlands. Veget Hist Archaeobot 17:93–103

Vuorela I, Lempiäinen T (1993) Palynological and palaeobotanical investigations in the area of the post-medieval Helsinki old town. Veget Hist Archaeobot 2:101–123

71 Wania A, Kühn I, Klotz S (2006) Plant richness patterns in agricultural and urban landscape in Central Germany – spatial gradients of species richness. Landscape and Urban Planning 75:97– 110

Wasylikowa K (1978) Early and Late Medieval plant Romains from Wawel Hill in Cracow (9– 10th to 15th century A.D.). Berichte der Deutschen Botanischen Geseuchaft 91(1):107–120.

Wasylikowa K, Cârciumaru M, Hajnalová E, Hartyányi BP, Pashkevich GA, Yanushevich ZV (1991) East–Central Europe. – In: Van Zeist W, Wasylikowa K & Behre KE (eds.), Progress in Old World palaeoethnobotany: 217–222, Balkema, Rotterdam.

Webb DA (1985) What are the criteria for presuming native status? Watsonia 15:231–236

Weber EF (1997) The alien flora of Europe: a taxonomic and biogeographic review. J.Veg. Sci. 8:565–572

Wieckowska M, Dörfler W, Kirleis W (2012) Vegetation and settlement history of the past 9000 years as recorded by lake deposits from Großer Eutiner See (Northern Germany). Rev Palaeobot Palynol 174:79–90

Wiethold J (1995) Plant remains from town-moats and cesspits of Medieval and Post-Medieval Kiel (Schleswig-Holstein, Germany). In: Kroll H Pasternak R (eds.) Res archaeobotanicae. 359–384, International Workgroup for Palaeoethnobotany. Proceedings of the nineth Symposium Kiel 1992

Wiethold J (2000) Macro-restes vegetaux imbibés et caronisés de la premier age du Fer provenant de fouille sauvetage de chantier ‘Les résidence du vignoble’ à Hattstatt– ‘Ziegelscheuer’ (Haut–Rhin). In: Dumont A, Treffort JM (eds.) Puits à eau protohistoriques (Bronze ancien et Hallstatt C). Afan Grand-Est, Strasbourg, SRA Alsace

Willcox (1996) Evidence for plant exploitation and vegetation history from three Early Neolithic pre-pottery sites on the Euphrates (Syria). Veget Hist Archaeobot 5:143–152

Willcox G (2012) Searching for the origins of arable weeds in the Near East. Veget Hist Archaeobot 21(2):163–167

Willerding U (1986) Zur Geschichte der Unkräuter Mitteleuropas. – Karl Wachholtz Verlag, Neumünster

Williamson M (1996) Biological invasions. Chapman and Hall, London

Williamson M, Dehnen-Schmutz K, Kühn I, Hill M, Klotz S, Milbau A, ... & Pyšek P (2009) The distribution of range sizes of native and alien plants in four European countries and the effects of residence time. Diversity and Distributions 15(1):158–166

Wilson JR, Richardson DM, Rouget M, Procheş Ş, Amis MA, Henderson L, Thuiller W (2007) Residence time and potential range: crucial considerations in modelling plant invasions. Diversity and Distributions 13(1):11–22

72 Zohary D, Hopf M, Weiss E (2012) Domestication of Plants in the Old World: The origin and spread of domesticated plants in Southwest Asia, Europe, and the Mediterranean Basin. Oxford University Press

Zohary M (1973) Geobotanical foundations of the Middle East. 2 Volumes. Gustav Fischer Verlag, StuttgartV

Paper I

Early to high medieval colonization and alluvial landscape transformation of the Labe valley (Czech Republic): evaluation of archaeological, pollen and macrofossil evidence

Radka Kozáková, Petr Pokorný, Věra Čulíková, Jan Mařík, Ivana Boháčová, Adéla Pokorná

Authorship statement

Paper I: Kozáková R, Pokorný P, Mařík J, Čulíková V, Boháčová I, Pokorná A (2014) Early to high medieval colonization and alluvial landscape transformation of the Labe valley (Czech Republic): evaluation of archaeological, pollen and macrofossil evidence. Veget Hist Archaeobot 23:701–718

 data collection: 15%  writing: 10% ...... Early to high medieval colonization and alluvial landscape transformation of the Labe valley (Czech Republic): evaluation of archaeological, pollen and macrofossil evidence

75 Radka Kozáková, Petr Pokorný, Věra Čulíková, Jan Mařík, Ivana Boháčová, Adéla Pokorná

Abstract

In the High Middle Ages, a wave of landscape transformation which originated in western Europe swept across the east-central part of the subcontinent. In the Czech Republic, this happened during the 13th century and it had the same environmental attributes as in the rest of Europe–a considerable increase in population, vast deforestation resulting in a rapid increase in soil erosion, irreversible changes in forest species composition and overall formation of cultural landscape. In the Czech Republic, the dynamics of such a radical change are poorly understood because it would require detailed archaeological, historical and palaeoecological insight into developments during the Early Middle Ages–a demand that is mostly not met. The aim of this paper is to fill in this gap. Archaeological and historical data from three early medieval strongholds located in central Bohemia, at Libice nad Cidlinou, Stará Boleslav and Hradišťko, are summarized and evaluated. The first two sites represent well-known political and religious centres of the early Czech state in the tenth to 11th centuries, while the last was of secondary importance. These archaeological sites have radiocarbon dated pollen and plant macrofossil evidence from oxbow sedimentary sequences which are situated in the immediate vicinity of the strongholds. The issue of fluvial transport of pollen and macrofossils is also discussed. Both pollen and macrofossil data from Hradišťko show surprisingly small impact of the stronghold on the forested alluvial environment. The vicinity of Stará Boleslav was intensively affected by human activity only during the later 11th century. It has not been possible to trace any impact of the foundation of the stronghold at Libice nad Cidlinou on the landscape. Medieval landscape change began before the 13th century in some places, as shown by the data from Stará Boleslav.

Introduction

The landscape transformation that occurred at the beginning of High Middle Ages in western and central Europe is reflected in many pollen diagrams (Rösch 2000; Ralska-Jasiewiczowa et al. 2004; Brown and Pluskowski 2011; Giesecke et al. 2011; Wieckowska et al. 2012). This event, with such striking and widespread effects, was already recognised by Franz Firbas who distinguished a biostratigraphic period called the Jungsubatlantikum (Later Subatlantic) (Firbas 1949). In the Czech Republic, this radical landscape transformation occurred around the mid 13th century AD (Klápště 2012). At that time, the population grew substantially, probably in connection with the process of massive colonization. Large numbers of new inhabitants created the need to establish an appropriate pattern of settlements, which has survived with only little change until today. In the course of a single century, increased human activity formed a cultural landscape that was much more similar to the modern landscape than to the landscape of, for instance, 9th century (Opravil 1983; Sádlo et al. 2008).

We can consider the Early Middle Ages as a transitional period when all the later transformation must have started. Archaeological data from the Czech Republic has shown a gradual trend of concentration and extension of populated areas as early as the 6th century as a result of increasing population following its great decline in the Migration Period (Klápště1994;

76 Kuna and Profantová eds. 2005). In combination with early written sources originating from the beginning of the 10th century, archaeological data provide a rather comprehensive picture of political organization and the basic structure of the emergence of the early medieval Czech state (Sláma 1988; Boháčová 2011). However, our knowledge of economic structure or cultural landscape development remains rather limited. The pollen data mostly give us an idea about landscape characteristic of a wider region, as the sampling sites are usually located in wetlands and thus far from the populated areas. Therefore, we are not exactly familiar with the situation of the cultural landscape before the 13th century, and whether there was an overall progressive increase in human activities, or whether isolated patches of highly altered cultural landscape only existed around the important population centres that gradually grew and later merged into a continuous agricultural landscape.

Fig. 1 Czech Republic with the indications of early medieval centres mentioned in the text (dots) and our sites (squares): 1 Stará Boleslav, 2 Libice nad Cidlinou, 3 Hradišťko

This paper focuses on the above-mentioned issues. As a background, archaeological and historical data from three early medieval strongholds situated on alluvial lowlands of the river Labe floodplain within the old settlement zone of the Czech Republic, Stará Boleslav, Libice nad Cidlinou and Hradišťko, are summarized (Fig. 1). The first two sites played an important role in the formation processes of the early medieval Czech state during the tenth and 11th centuries. The seats of local rulers were based at Stará Boleslav and Libice nad Cidlinou, and these strongholds were also centres of increasing Christianity, and overall central places with relatively high population densities and levels of craft production (Boháčová ed. 2003; Mařík 2009). The stronghold at Hradišťko is a representative of a later feature of the settlement structure. It is smaller in size, shows limited evidence of craft production and was, of course, probably inhabited by a smaller group of people (see below). Sites of this type filled in the spaces between important sites of the first-rank (Mařík 2013).

77 While searching for archaeobotanical evidence, we were able to obtain pollen and macrofossils from the sediments of old oxbow lakes in old river channels that are located in the immediate vicinity of the strongholds (Figs. 2, 3, 4). In the course of ongoing archaeological research in Stará Boleslav and Libice nad Cidlinou, numerous palaeoenvironmental studies on pollen, plant macrofossils and bones have already been performed on samples from cultural layers (Čulíková 1999, 2003, 2006; Mlíkovský 2003). Unfortunately, the application of the data from cultural layers to the drawing of conclusions regarding landscape development is rather limited. In territory of the present Czech Republic, other important alluvial strongholds existed, especially the significant centres of the Great Moravian Empire founded during the 9th century: the sites of Mikulčice and Pohansko (Fig. 1; Macháček et al. 2007). Pollen and macrofossil data from these two strongholds were obtained from flood loams, subfossil soils and partly also from cultural layers (Svobodová 1990; Doláková et al. 2010). A complicated sedimenthology together with the absence of radiocarbon dates makes the comparison of those results with our data difficult. In the case of some off-site profiles, the absence of radiocarbon dates again makes their applicability for chronological purposes problematic (Svobodová 1990). Thus, due to the circumstances mentioned above, no really close analogies for the results presented here can be found in the present Czech Republic.

We believe that the comparison of historical, archaeological and palaeoecological data will allow us to determine the impact the strongholds exercised on their natural environment in great detail. Nevertheless, there exist certain limits to our data, especially in the case of pollen evidence. Pollen data from old oxbows require specific methodological approaches, which are discussed and taken into account in our study.

The following questions will also be asked and discussed in this article: (1) Is our pollen and macrofossil evidence sensitive enough to reflect settlement events; at least those that are attested by archaeological research? (2) Did the early medieval strongholds continue from some previous settlements on the sites or their close vicinity? (3) How did the foundation of the strongholds influence the natural environment in their surroundings? (4) Was the landscape surrounding the strongholds already significantly changed by human activity in early medieval times or did the radical landscape transformation occur only in the High Medieval Ages?

Fig. 2 Map showing the Stará Boleslav stronghold with the indications of early medieval archaeological finds and pollen and macrofossil profiles. Pollen profile Stará Boleslav 2 (grey star) was analyzed by Eva Břízová (1999)

Location and description of sites

Natural conditions

All the three studied strongholds are situated in the central Bohemian lowland that is significantly influenced by the river Labe (Fig. 1). The pre-Quaternary bedrock consists of Upper Cretaceous sediments, mostly claystones, siltstones and marlstones. The wide and shallow valley of the river Labe is filled with Quaternary sands and gravels that form fluvial terraces, of which the latest and at the same moment the lowest one is of early Holocene age (Dreslerová et al. 2004). All the observed sites are located on geomorphological relicts of these terraces in a close contact with the Holocene floodplain and with abandoned oxbows. In comparison with the rest of the Czech Republic, the region has a relatively high mean annual temperature (9–10 °C) and a relatively lower mean annual amount of rainfall (560–600 mm). The present day landscape is cultural with cultivated fields and little remaining woodland.

Fig. 3 Map showing the Libice nad Cidlinou stronghold with the indications of early medieval archaeological finds and pollen and macrofossil profile

Fig. 4 Map showing the Hradišťko stronghold with the indications of early medieval archaeological finds and pollen and macrofossil profile

80 Archaeological and historical background

Stará Boleslav

The stronghold was founded on a remnant of a fluvial terrace, which rises at least 2 m above the alluvium of the floodplain (Fig. 2). The area surrounding the stronghold is one of the regions which have been continuously inhabited since the late Neolithic period (4200–2200 cal BC). A combination of historical and archaeological sources dates the foundation of the Stará Boleslav stronghold to around AD 900 (Boháčová ed. 2003; Boháčová 2006). Stará Boleslav was a seat of the Přemyslids dynasty in the early development phases of the Czech state. The stronghold was situated at the edge of the area controlled by the Přemyslids at the crossing of the river Labe by the important road leading from central Bohemia to the north and northeast which formed a connection with the principal European network of roads.

The fortified part of the stronghold covered approximately 5 ha. The intensively populated outer bailey (11 ha) adjoined the fortified part to the east and no fortification has been found there so far (Fig. 2). According to archaeological evidence it seems that there was no early medieval settlement immediately prior to the 9th century, when the stronghold of Stará Boleslav was founded. Archaeological finds from earlier phases of the stronghold’s existence have clearly showed that the inhabitants of Stará Boleslav were of a rather exclusive social status (Boháčová 2003). At that time, the ruler’s power was demonstrated not only by the foundation of a church, but also by construction of a unique fortification wall.

Later, prior to the mid 11th century, Stará Boleslav became a primary religious centre of the Czech state, and one of the earliest canonries with appropriate basilica and two other churches were built there. Precisely at this period of time, archaeological evidence shows striking changes in organization and use of the inhabited areas, such as new burial grounds around the churches or a paving of previously inhabited areas in the stronghold’s centre. At the same time, local craft production significantly increased, especially local pottery and floor tiles, and processing of silver, iron and non-ferrous metals.

No later than the 12th century, Stará Boleslav started to lose its importance. Archaeological evidence from this time has revealed a decreased rate of formation of cultural layers, and a lack of maintenance of the infrastructure as well as few repairs to the above- mentioned stone fortification. Continuity in deposition of cultural layers ends some time during the first half of the 13th century. Archaeological finds from the beginning of the High Middle Ages in the first half of the 13th century are rather rare, which means that development of the stronghold of Stará Boleslav did not continue. Even though the stronghold was still inhabited, its central part was not rebuilt or reconstructed.

In its broader vicinity, the early medieval settlement followed the edges of the river floodplain on the left bank of the river Labe, between the city of Prague (Praha) and Stará Boleslav, also on the long distance trade road, mentioned above. During the course of eleventh and 12th centuries, new domains belonging to several Prague religious institutions were founded in the area between Prague and Stará Boleslav. On the other hand, areas adjoining the

81 stronghold to the east were rather thinly populated and only a little farmed in the early medieval period.

Libice nad Cidlinou

The stronghold of Libice nad Cidlinou (from now on referred to as Libice) was founded on two island-like remnants of a fluvial terrace with a total area of 24 ha on which stand the inner and outer baileys, or courtyards. The terrace topography was created by the erosion of the rivers Labe and Cidlina (Fig. 3).

The earliest traces of settlement on these two areas of fluvial terraces are dated to the middle Bronze Age (ca. 1600–1200 cal. BC). The earliest early medieval settlement at the site can be associated with the arrival of the Slavonic ethnic group in the course of the sixth and 7th centuries. At this time, archaeological data indicate that an unfortified smaller settlement stood on the outer bailey. A find of belt fittings discovered in the inner bailey suggests settlement continuity during the seventh and 8th centuries. A distinct increase in settlement activities at the site is dated to the late 9th century, when large burial grounds were established in the inner bailey and its close surroundings (Fig. 3). Finds of graves containing jewellery and weapons definitely corroborate the presence of higher social strata. The major change in settlement intensity and structure at the end of the 9th century can be interpreted as a foundation phase of the stronghold. The position of the stronghold of Libice at the crossing point of two long- distance transport and trade routes leading northwards to the Silesia and eastwards across Moravia to eastern Europe was very beneficial to it. From the late 9th century onwards, a new settlement zone appeared on the northern bank of the river Cidlina that stood apart from the central areas of the stronghold with the inner and outer baileys. Another important change in the stronghold’s history occurred in the mid 10th century when a stone church was built in the inner bailey. At the same time, a fortification was built encircling the inner and outer baileys. The existence of an earlier fortification dated to the late 9th century is also plausible, but no absolutely reliable evidence has survived. In the second half of the 13th century, Libice appeared for the first time in written sources in connection with the Slavníck dynasty, a noble family who ruled the stronghold until AD 995. During the 11th century, there is significantly less settlement evidence at the stronghold, both in the fortified centre and its vicinity. Settlements on the north bank of the river Cidlina were abandoned and the fortified section of the stronghold was used only as a burial ground. The inner bailey was also abandoned and a new village, already in High Medieval style, was founded in the outer bailey. In the year 1130, the stronghold of Libice appeared in written sources for the last time as a fortified centre (oppidum).

The fact that the stronghold of Libice was an important early medieval centre is reflected in both archaeological and written sources. Among crafts, there is archaeological evidence for iron production, jewellery workshops where gold and silver were used and textile production reflected by numerous finds of whorls. The mint that according to the written sources must have existed at the stronghold at the end of the 10th century has remained so far archaeologically undetected. Economic models suggest that the main resources such as crops, fuel wood, timber and stones for various constructions, which were needed for life at the stronghold, could mostly

82 be obtained from the surrounding land up to a maximum radius of 4 km (Mařík 2009). The fact that only a few other settlements have been found in this area supports this presumption.

Hradišťko

This settlement is situated on the east bank of the river Labe on a sand and gravel fluvial terrace remnant, approximately 7.5 km south of Libice (Fig. 4). The total area of the settlement covered 5.2 ha. In the eastern part of the site, apparent remnants of fortification, which encircled part of the settlement (2.4 ha), are still visible in the fields. During the surface surveying, two distinct concentrations of the early medieval pottery were observed, one within the fortified area, while the other was beyond the fortification. Tentatively, the foundation of this settlement can be dated to the transition from the tenth to the 11th century, and it was probably abandoned by the 12th century. Even though the stronghold of Hradištko is relatively less known, the site is interesting from the point of view of the regional settlement structure development. One of the long-distance routes running towards Libice and Stará Boleslav passes by the Hradišťko stronghold. Another important regional centre was the site of Kolín that is situated approximately 11.5 km to the south of the stronghold of Libice (Fig. 1). It seems that Hradišťko and two other similar settlements located between Libice and Kolín were probably founded to meet the demands for more intensive agricultural production. They may also represent the result of the expanding state administration of the young Czech state.

Table 1 Overview of the radiocarbon dates. The measurements were made in the radiocarbon laboratory in Poznan, Poland (AMS method) and in the CRL radiocarbon laboratory in Prague (decay counting method)

14C age Cal. age AD Site Depth (cm) Type Lab. code Material (years BP) (1 σ-range) Stará Boleslav 60–65 AMS Poz-26551 1,010 ± 30 1,013 ± 17 Fruits of Carex Stará Boleslav 75 Conv. 7161 1,166 ± 82 850 ± 96 Wood Stará Boleslav 100–110 Conv. 7160 1,559 ± 87 488 ± 85 Wood Libice 26–29 AMS Poz-33152 945 ± 35 1,089 ± 48 Fruits of Carex Libice 71–74 AMS Poz-33153 1,070 ± 30 955 ± 42 Fruits of Carex, Sparganium Libice 113–116 AMS Poz-33150 1,325 ± 35 702 ± 42 Fruits of Carex, Potamogeton Hradišťko 107 AMS Poz-29398 760 ± 30 1,249 ± 19 Bark Hradišťko 140 AMS Poz-29399 1,160 ± 30 871 ± 58 Fragment of fruit Hradišťko 161 AMS Poz-33146 1,170 ± 35 858 ± 60 Twig Hradišťko 182 AMS Poz-29400 1,230 ± 30 781 ± 62 Twig

Demographic estimates

Assessment of the numbers of people living in the early medieval strongholds remains rather problematic. No written sources that could be used for such demographic estimations have survived and thus the estimates can only be based on archaeological data. Due to the rather fragmentary character of archaeological evidence, only probable figures with quite wide error ranges can be predicted.

For the demographic estimates, the extent and structure of burial grounds, size of houses or area of the settlement (Hassan 1978) can be used as further indicators. In the case of early medieval strongholds, the data from burial grounds are most suitable, however these have only rarely been completely excavated. For the sites discussed in this paper, the only sufficient data comes from Libice, where a rather dense network of archaeological test pits has enabled the

83 reconstruction of the most probable extent of burial grounds there which have not been excavated so far. On the basis of this reconstruction and the data obtained from excavated burials, the number of inhabitants of Libice in the 10th century can be estimated as 600–900 (Mařík 2009). In the case of the other two sites, Stará Boleslav and Hradišťko, we are lacking either a sufficient number of excavated burials or we do not know the extent of the burial grounds. In the case of Stará Boleslav, the size of houses cannot be used as an indicator because the structure of the inhabited area is unknown. Therefore, the last approach, the least accurate one, based on the size of the populated area can be applied. This approach is usually used for population size estimates for ancient towns in the eastern Mediterranean (Kolb 2005). However, these towns were of various structures and fulfilled very different functions and so the regularly applied figure of 200–300 inhabitants per hectare applied there cannot be simply transposed to the central Bohemian region. Population estimates from Libice (total area 24 ha) indicate a figure of 25–37.5 inhabitants/ha. Similar results (37.8 inhabitants/ha) have been obtained in demographic analyses of burials that were performed on the early medieval stronghold of Mikulčice (Stloukal and Vyhnánek 1976) as well as the early medieval town of Haithab in the northern Germany (42 inhabitants/ha; Steuer 1984). By using the figure from Libice we can estimate 400–650 inhabitants for Stará Boleslav and 90–150 for Hradišťko.

Samples for pollen analysis were prepared according to Faegri and Iversen (1989). Sediment was digested in 10% KOH, carbonates were removed by applying concentrated HCl, silicates dissolved in concentrated HF and most of the organic material removed during acetolysis, using acetic anhydride and sulphuric acid in a 1:9 ratio. Usually, more than 500 grains were counted per sample. Pollen types were defined according to Punt (1980), Reille (1992) and Beug (2004). Pollen grains were well preserved in most of the samples from organic as well as fluvial sediments. In the case of the site of Libice, no pollen samples in the interval between 44 and 72 cm were analysed, because the pollen in this section was strongly degraded or completely absent. Similarly, in the macrofossil diagram, samples between 37 and 67 cm were excluded for their low content of macrofossils (0-16).

The pollen diagrams are based on a total terrestrial pollen sum from which the following taxa were excluded, Alnus, Salix, all aquatic and wetland herbs and all non-pollen objects. Microscopic charcoal particles were counted using the point-count method (Tolonen 1986). The curves of all pollen and non-pollen objects that were excluded from the sum were calculated as percentages of the total terrestrial pollen sum. Our data from Stará Boleslav are compared with an earlier research, which was sampled from the same oxbow as our pollen site, and is referred to here as Stará Boleslav 2 (Fig. 2; Břízová 1999).

Fig. 5 Percentage pollen diagram of selected taxa for the site of Stará Boleslav, analyzed by P. Pokorný. Grey silhouettes represent x 10 exaggeration of the scale. Arrow indicates expected date of foundation of the stronghold. Lithology: 1 fluvial sand, 2 organic sediment with sand (grains up to 1 mm), 3 organic sediment with wood and plant macrofossils, 4 organic sediment with a certain amount of flood loam

Fig. 5 continued

For macrofossil analysis, the sampling intervals were 3 cm (Hradištko and Libice) and 5 cm (Stará Boleslav). The sediment was wet sieved on 0.25 and 0.5 mm meshes. Plant diaspores and vegetative parts were then picked out under a dissecting microscope.

Pollen and macrofossil diagrams were plotted using Tilia 1.7.16 (Grimm 1992). Diagrams were zoned subjectively according to distinct changes in presence and abundance of ecologically important taxa.

85 The radiocarbon dates for the profiles are presented in Table 1 with all complementary information.

Fig. 6 Absolute macrofossil diagram for the site of Stará Boleslav, analyzed by A. Pokorná and P. Pokorný. Expected date of foundation of the stronghold is indicated by the arrow. Most common and indicative taxa selected from the total sum of 52. x-axes, count of macrofossils in 350 cm3 of sediment

Archaeological research

The site of Stará Boleslav was intensively explored in the course of a systematic rescue archaeological research programme which was conducted especially in the years 1988–2001, when some representative archaeological sites were excavated both in the centre and in the adjacent areas of the stronghold (Boháčová 2003, 2006, 2011). Stratification of the sediments showed several early medieval horizons which could be dated at least to a time interval on the basis of presence of the Prague pottery sequence, whose chronology was defined during the excavations at Pražský hrad (Prague castle) by available dendrochronological dates (Boháčová 2001). Also, written sources from the tenth and 11th centuries provided some dating information because several medieval authors mentioned at least approximate dates of some events which can be detected archeologically, such as construction of the basilica in AD 1039– 1046, or the unique stone fortification of the 930s.

The stronghold of Libice has been archaeologically surveyed continuously since 1949. Archaeological test pits cover approximately 5% of the total fortified area and a further 15 000 m2 have been excavated in the adjoining village, where the majority of building activities have been accompanied by rescue archaeological fieldwork. Archaeological research conducted on the inner bailey occurred mainly in 1949–1974 and focused predominantly on the church and adjoining burial ground. Since 2010, the entire inner bailey has been studied by non-destructive methods such as geophysical survey, surface collections or aerial photographic interpretation.

As the stronghold of Hradišťko is omitted from the written sources, we must rely solely on archaeological evidence. The earliest archaeological test pits excavated at the end of the 19th century definitely dated the site to the early medieval period. Several settlement features from

86 the tenth to 11th centuries were excavated in 1990. Later analytical surface collections have enabled estimation of the total extent of the settlement (Fig. 4).

Results

Stará Boleslav

Pollen data

SB P1 (AD 399-1026) According to the calibrated radiocarbon chronology, the sedimentary record starts around AD 400 (Fig. 5). The composition of pollen spectra is steady, with a prevalence of Quercus. Even though decreases in Quercus are noticeable, but only in several single layers, in general no profound decrease in arboreal pollen is evident during this zone. There was an area of cultural landscape indicated by low and discontinuous pollen curves of cereals, ruderals and weeds, and this was encircled by the forested alluvium surrounding our pollen site. Initially, a pool had existed in the old oxbow (as shown by pollen of Potamogeton, Myriophylum, and Nuphar) but during the first half of this zone the site became a wetland. In the upper part of this zone, the stronghold’s foundation is supposed to have taken place.

SB P2 (AD 1026-1129) A decrease in Pinus and an abrupt increase in indicators of human activities such as Triticum type, Secale cereale, Artemisia, Urtica, Brassicaceae, Rubiaceae, Rumex acetosa-type and Trifolium-type is clearly visible. Quercus forests were still dominant in the region during this stage.

SB P3 (AD 1129-1275) Radical vegetation changes mark the beginning of this zone. Quercus rapidly declines together with an ongoing decline in Pinus. Quercus was partly replaced by Carpinus, which grows faster and regenerates well from stumps after cutting or grazing. Eutrophication very probably represents the cause of the distinct regeneration of Acer and Ulmus in the alluvium. For the same reason, Sambucus nigra, Urtica, Solanum dulcamara, Symphytum or Filipendula expanded. The dominant local vegetation community was alder carr, and local Alnus glutinosa stand were obviously not cut for timber.

SB P4 (AD 1275-1406) The most recent part of the pollen profile is connected with the change in sedimentation. Instead of organic sediments, flood loams gradually filled in the old oxbow. One of the major vegetation changes represents the expansion of Pinus that, above all, replaced Carpinus and Quercus. From the beginning of this zone, pollen of cereals increased, and Centaurea cyanus, a typical weed of the High Middle Ages occurred for the first time. Artemisia was very common as a ruderal and could indicate abandonment of areas of dry ground or, at least, their less intensive use.

Macrofossils

SB M1 (AD 399-1000) Finds of Potamogeton and Nuphar lutea together with larval cases of Trichoptera and also Alisma plantago-aquatica and Oenanthe aquatica corroborate the existence of shallow stagnant waters in the oxbow that diminished in the second half of this zone (Fig. 6). Alnus was present in waterlogged zones around the oxbow. Macrofossils of

87 Quercus, Tilia, and Betula together with shrubs represented by Cornus sanguinea and C. mas indicate that a certain part of the alluvium was cavered by mixed deciduous woodland.

SB M2 (AD 1000-1246) From this zone onwards, Quercus disappears from the macrofossil assemblage and Carpinus is relatively frequent. A rapid increase in Alnus shows its expansion on the alluvium. Several taxa that expanded during this zone, especially Persicaria cf. hydropiper and Urtica dioica, indicate eutrophication. Human impact is further evident from the finds of Xanthium strumarium and Polygonum aviculare.

SB M3 (AD 1246-1406) Alnus was reduced, and ruderal vegetation grew around the site, which is evident from the numerous macrofossils of Sambucus nigra.

Libice nad Cidlinou

Pollen data

LC P1 (AD 693-936) The second half of this zone corresponds to the foundation of the stronghold according to the absolute chronological data (Fig. 7). Arboreal pollen is slightly over 50%, which is the lowest percentage in comparison with the other two sites in a similar period of time. In the waterlogged land on the alluvium, Salix and Alnus were present, while mixed deciduous woodland with dominant Quercus occupied drier places. In the second half of the zone, some decrease in Quercus is evident. The pollen of shrubs represented by Prunus-type, Sorbus-type and Rhamnus is quite frequent considering the relatively low pollen production and dispersal of these taxa, and represents scrub which could have grown as wayside or field edge vegetation around the stronghold, while others such as Frangula alnus and Viburnum opulus would have grown in wet places around the oxbow. A continuous curve of Potamogeton shows the existence of a pool at the site in the earlier half of the zone. A cultural landscape on drier land surrounding the oxbow is indicated by plentiful pollen of cereals, ruderals and weeds. Occurrence in the sediment of the ova of Trichuris, an intestinal parasite of mammals as well as humans, is a sign of faecal pollution. A rich pollen spectrum of herbs reflects pastures and meadows.

LC P2 (AD 936-1118) This zone is marked by an abrupt change in the lithology that can be dated to the end of the 10th century (Fig. 7). Organic sediment was replaced by fine and impermeable flood loams, and pollen was poorly preserved or completely absent between 44 and 72 cm. Such a change was probably the result of the extensive loss of woodland in regions in the upper basin of the river Labe. Quercus pollen counts are radically reduced and Alnus also distinctly decreases. A rapid increase in Poaceae pollen very probably indicates expansion of reedbeds in riverine wetlands. Also, Urtica, Cyperaceae, S. dulcamara, Humulus lupulus and other wetland taxa become more common. The increase in crops, ruderals and weeds is not very significant. Increased human impact is more clearly reflected by increased charcoal percentages or numerous pollen of Trifolium repens-type showing increasing grassland.

Fig. 7 Percentage pollen diagram of selected taxa for the site of Libice nad Cidlinou, analyzed by R. Koza´kova´. Grey silhouettes represent 910 exaggeration of the scale. Expected date of foundation of the stronghold is indicated by the arrow. Empty space between 44 and 72 cm represents section with no or badly preserved pollen. Lithology: 1 sand, 2 organic sediment with numerous plant macrofossils and a very small amount of fine silt; 3 flood loam, grey, finegrained, plastic and homogenous

Fig. 7 continued

Macrofossils

LC M1 (AD 693-936) Aquatic taxa indicate the presence of mesotrophic or naturally eutrophic water at the site (Fig. 8). All the present aquatic plants and especially green algae of the genus Chara are sensitive to eutrophication. Marsh plants are represented by several taxa indicating shallow still water, Alisma plantago-aquatica, Oenathe aquatica and Sagitaria sagittifolia. A number of other wetland taxa formed species rich bank and sedge communities. Ruderal bank

89 Fig. 8 Absolute macrofossil diagram for the site of Libice nad Cidlinou, analyzed by V. C ˇ ulı´kova´. Most common and indicative taxa selected from the total sum of 130. x-axes, count of macrofossils in 210 cm3 of sediment. Expected date of foundation of the stronghold is indicated by the arrow. Empty space between 37 and 67 cm represents section with very low macrofossil sums (0–16). Taxa included in Carex sp. div. (fruit and utricle) are Carex acuta, C. hirta, C. pseudocyperus, C. vesicaria. Taxa included in Rumex sp. div. (fruit) are Rumex cf. aquaticus/hydrolapathum, R. crispus/obtusifolius, R. hydrolapathum, R. maritimus, R. cf. obtusifolius, R. cf. sanguineus, R. conglomeratus/obtusifolius. Taxa included in Chenopodium sp. div. (seed) are C. album, C. ficifolium, C. hybridum, C. polyspermum

90 vegetation is shown by Urtica dioica, Polygonum hydropiper and various Rumex taxa. Macrofossils of Quercus are rare compared with the other two sites (ESM). The only locally common trees were Alnus and Salix. A few macrofossils indicate the cultivation of Panicum miliaceum, Triticum and Cannabis sativa (Fig. 8, ESM). Local cultivation is further reflected by relatively numerous finds of cereal weeds. Among ruderal and weed communities, annual nitrophilous plants such as Atriplex and Chenopodium, perennial nitrophilous taxa such as Aegopodium podagraria and Silene latifolia, annual weeds often associated with root crops such as Solanum nigrum, Galeopsis, Hyoscyamus niger, Lamium purpureum and Echinochloa crus-galli, and a typical indictor of trampled vegetation, Polygonum aviculare, are present (Fig. 8, ESM).

LC M2 (AD 936-1118) The most distinct difference from Zone LC M1 zone is caused by the distinct change in sedimentation (Fig. 8). In the interval 37-72 cm with poorly preserved or absent pollen, the number of macrofossils is also low, with 0-16 finds, and therefore these samples were excluded. In the rest of the zone, the total sum of macrofossils is only one tenth of that in Zone LC M1. Therefore, the overall decrease in the number of identified taxa is caused by both decrease in the total sum of macrofossils and the effects of human activities on the site. Macrofossils from this zone support the pollen results showing the likely disappearance of Alnus from the site.

Hradišťko

Pollen data

HR P1 (AD 705-796) The sediment in the lowest section of the profile is organic combined with fine flood loams and it contains layers with numerous plant macrofossils (Fig. 9). Arboreal pollen generally reaches over 80%. Species-rich mixed deciduous woods where Quercus prevailed were common on alluvium and probably also in the whole region. Water-logged places on alluvium were dominated by alder carr. The oxbow was not yet filled with sediments, as shown by numerous pollen of aquatic plants such as Potamogeton and Nuphar. Solanum dulcamara, Filipendula and especially Urtica pollen indicate eutrophic wetland vegetation. The landscape surrounding the river was cultural with the same characteristics as in the following phase.

HR P2 (AD 796-1349) The sediment is similar in character to the previous zone. The foundation of the stronghold corresponds approximately to the middle of this zone, according to the absolute chronological data (Fig. 9). A gradual reduction in arboreal pollen is evident, but the taxa composition of mixed deciduous woodland was rich and steady, with Quercus, Corylus, Carpinus, Tilia, Ulmus and Pinus. Alnus and Salix grew in waterlogged places on the alluvium and the pool in the oxbow still existed. Human impact is directly shown by pollen of cereals, mostly Secale cereale, and Cannabis sativa was also detected. Continuous pollen curves of Plantago lanceolata and Rumex acetosa-type as well as pollen of numerous herbs suggest pastures and meadows that could have been extensive.

Fig. 9 Percentage pollen diagram of selected taxa for the site of Hradisˇtˇko, analyzed by R. Kozaáková. Grey silhouettes represent 910 exaggeration of the scale. Expected date of foundation of the stronghold is indicated by the arrow. Lithology: 1 sand, 2 organic sediment with numerous plant macrofossils and a very small amount of fine silt and sand (grains up to 0.5 mm), 3 organic sediment with plant macrofossils and a certain amount of flood loam, 4 flood loam, grey-ochre, containing rootlets and spots of ferrous precipitates

Fig. 9 continued

HR P3 (AD 1349-1572) Radiocarbon dating suggests that the radical vegetation change detected at the beginning of this zone can be connected with the massive clearance of woodland in the 13th century and later (Fig. 9). Most of the oak woods were cut down in the region and Alnus also nearly disappeared from the flood plain. Even Abies and Fagus decreased, which means that the clearance of woodland was overall and also affected the uplands about 20 km away which had suitable ecological conditions for them to grow. As a consequence, from 85 cm masses of eroded soil started to accumulate in the oxbow in the

92 154. x-axes, count of macrofossils in 210 cm3 of sediment. Expected date of foundation of the stronghold is indicated by the arrow. Taxa included in Carex sp. div. (fruit and utricle) are Carex cf. acuta, C. sp. cespitosa, C. x dioica-type, C. echinata, C. flava agg., C. hirta, C. nigra, C. cf. otrubae. Taxa included in Rumex sp. div. (fruit) are Rumex cf. crispus, R. cf. obtusifolius, Rumex sp. Taxa included in Chenopodium sp. div. (seed) are C. album, C. ficifolium, C. cf. glaucum, C. hybridum, C. polyspermum

Fig. 10 Absolute macrofossil diagram for the site of Hradišťko, analyzed by V. Čulíková. Most common and indicative taxa selected from the total sum of

form of flood loams. A substantial increase in Poaceae very probably shows the expansion of reedbeds on this new substrate. An increase in cultivated crops shows a rather intensified agriculture and a rapid increase in weeds and ruderals shows overall development of a cultural landscape. Also, grazing intensified, which is especially reflected by a significant increase in Trifolium repens-type.

HR P4 (AD 1572-1799) Original mixed deciduous woodland was mainly replaced by Pinus, but some regeneration of Quercus, Tilia and Carpinus is also evident. The increase in Picea very probably reflects modern forest plantations.

Macrofossils

HR M1 (AD 705-1051) Macrofossils of numerous aquatic plants indicate the presence of mesotrophic still water at this stage (Fig. 10, ESM). In the waterlogged vicinity of the oxbow, species-rich alder carr occurred, in which Alnus was the most common but Salix, Fraxinus and Prunus padus were also present. Drier parts of alluvium were still wooded as indicated by the macrofossils of Quercus, Tilia, Corylus, Crataegus and Carpinus (Fig. 10, ESM). The cultural component of the macrofossil assemblages is more or less the same as in the following stage, except that the finds are less frequent.

HR M2 (AD 1051-1349) The beginning of this zone corresponds to the foundation of the stronghold (Fig. 10). A decrease in Alnus macrofossils probably shows some local reduction of alder trees. Aquatic plants also decrease, which together with the simultaneous increase in Alisma plantago-aquatica and Oenanthe aquatica, could show that the water level was decreasing (Fig. 10). A number of ruderals and weeds occur in this zone, especially Chenopodium album, Polygonum aviculare, Stellaria media, Barbarea vulgaris, Galeopsis pubescens/tetrahit, Solanum nigrum and Fallopia convolvulus, reflecting local ruderalisation as well as human impact in the vicinity (Fig. 10, ESM). The presence of cereal crops is directly indicated by Panicum miliaceum macrofossils.

HR M3 (AD 1349-1506) This zone can clearly be correlated with a particular zone of the pollen diagram (HR P3 in Fig. 9). Alnus glutinosa as well as other trees disappeared from the site. A distinct decrease in the number of marsh plants and the dominant taxa, especially Carex hirta and Juncus, indicate human activities (ESM). Frequent occurrence of cereal weeds signifies extension of cultivated fields and fallow land around the site.

HR M4 (AD 1506-1799) This part of the profile was extremely poor in macrofossils. One exception represents the layer at 58 cm with enormous amount of Potamogeton macrofossils (Fig. 10), which were probably transported fluvially.

Discussion

Fluvial transport of pollen grains and macrofossils?

106 It is widely accepted among pollen analysts that most of the pollen and spores entering alluvial sedimentary basins are fluvially transported from river catchments (Peck 1973; Bonny 1978; Pennington 1979). Recently, Brown et al. (2007) performed an extensive modern analogue study of this process, demonstrating how important this knowledge is for pollen-based reconstructions of cultural landscapes. Our sedimentary records originated in alluvial oxbow sub-environments that were undoubtedly influenced by fluvial sedimentation. The evidence for this is the regular, though variable, occurrence of fine silt in the studied profiles.

Using the present state of the data, it is unfortunately impossible to directly study the quantitative relationships between the airborne and waterborne components of the pollen spectra (both terms used by Brown et al. 2007). Nevertheless, we may assume that some unknown proportion of pollen grains in our pollen slides was fluvially transported.

The result of these considerations may be a cautious approach to the effort to compare local settlement events, based on interpretation of archaeological evidence, with chronologically correlated pollen spectra. Our insight into local conditions may be blurred by a long-distance transported waterborne pollen component. In other words, we must take into account the serious possibility that a pollen signal may reflect changes in local environments to a lesser degree than in the case of "classical" pollen analytical sites such as lakes and peat bogs outside large river alluvia.

The above-mentioned considerations may count also for the macrofossils. Although these have different sedimentation characteristics (in physical sense), it is equally plausible that some of them may have been brought in by running water of the major river channel and do not have local origins.

Being aware of the interpretation limits of the data from a fluvial environment, we have to stress that our sites are old oxbows that were almost completely separated from the river and thus from the major fluvial sedimentation zone at the time of their formation. Thanks to this fact the inflow of fluvially transported silt into the sedimentary swquences was very low in the deeper parts of our profiles and greater only in the upper sections. In deeper parts of the three profiles the sediments are nearly completely organic (Figs. 2, 3, 4) and such a sedimentary sequence we consider to reflect predominantly local conditions. In the case of Hradišťko and Stará Boleslav, this organic section covers the crucial period of the Early Middle Ages, when the foundation and existence of the strongholds is recorded by archaeological methods. Only in the case of Libice, where flood loams sedimented already during the Early Middle Ages, does the tracing of the local environmental development around the stronghold remain problematic. The distinct change in the lithology in the upper parts of our profiles (number 4 in lithology columns, Figs. 5, 7, 9) certainly indicates an abrupt increase in the proportion of fluvially transported pollen and macrofossils. Except for the site of Libice, this process only occurred in the High Middle Ages, when the landscape transformation was already extensive and, thus, the requirement to detect local pollen spectra becomes less important for us. Nevertheless, when commenting on the landscape changes in the High Middle Ages we must be aware of the fact that our pollen signal reflects a larger spatial scale than the local one.

107

Stará Boleslav

Before the stronghold was founded, wooded alluvium without any settlements dominated the river valley. The first clearly evident colonization event is reflected by a distinct fall in the Pinus pollen percentage at the start of Local Pollen Zone SB P2 (Fig. 5). This can probably be connected with clearance of woodland on the colonized highest fluvial terrace, located outside the flooded zone, which was rather dry and thus suitable for pines. At the same time, woodland clearance and settlement facilitated the growth of Artemisia. Unfortunately, radiocarbon chronology (Fig. 5, Table 1) does not allow us to distinguish exactly whether the beginning of Local Pollen Zone SB P2 relates to the original foundation of the fortified settlement at Boleslav around AD 900, or if it can be connected with the later foundation of the basilica (church) in 1039–1046. The former is, however, more likely. In any case, this first significant phase of human activity seems to have had a rather local extent, because it did not yet affect the oak woods.

Extensive landscape change occurred only later and is marked by the beginning of Local Pollen Zone SB P3 (Fig. 5). Based on absolute radiocarbon chronology, this event can be dated to the 11th century and thus this section of the pollen diagram can be better connected with the foundation of the canonry, which certainly represented a significant increase in the economic and agricultural development of the site. In the 12th century, the importance of Stará Boleslav as a political, religious and representative centre was already gone (Boháčová ed. 2003). Basically, this decline is not reflected in pollen and macrofossil data. Only a distinct increase of Alnus, Salix and Acer can be observed in the pollen diagram, suggesting that the alluvial zone was abandoned and not exploited. Apart from this local vegetation change, the landscape in the region became more intensively used.

In the pollen diagram Stará Boleslav 2, from another site situated in the same oxbow less than 1 km away (Fig. 2), the general trends are similar. The prevalence of alluvial woodlands before and even after the expected time of the foundation of the stronghold is evident. Compared to our pollen data, the initial woodland composition around the site Stará Boleslav 2 was different, with a higher proportion of Alnus and Salix and a lower proportion of Pinus and Quercus (Břízová 1999). Such a result reflects the strong effect of pollen rain from the local vegetation. Both pollen diagrams seem to corroborate the assumption that the profound landscape transformation that is elsewhere connected only with the 13th century in the Czech lands, occurred in the region surrounding Stará Boleslav no later than during the 12th century. Even though the reasons for such an earlier landscape change in the Stará Boleslav region cannot be precisely distinguished, its relative proximity to Prague, the real centre of the state (Fig. 1), must be taken into consideration.

Libice

Even before the foundation of the stronghold, the vicinity of this site was mostly not wooded, probably reflecting the human impact of the previous settlement phase. The foundation of the stronghold was quickly followed by a distinct vegetation change, marked especially by a

108 decrease in trees that comprised the major woodland, Quercus, Tilia, Carpinus, Ulmus and Corylus. Also, Alnus and Salix on the alluvium were cut down. Simultaneously, the sediment type rapidly changed from predominantly organic into fine loams, probably as a consequence of the erosion following clearance of woodland in the region situated upstream. Such an evident correlation between vegetation and sediment change was not observed at the other two sites (Figs. 5, 9). Also the timing of the sedimentation change is different at Libice; this can be explained by the different sedimentation histories of the two rivers, the Cidlina at Libice and the Labe at the other two sites. Considering such an extreme change in sedimentation, we have to assume that there was also a relative increase in fluvially transported pollen, and so an increase in the long-distance fluvial component of the pollen spectra. If most of the pollen recorded in the upper part of the pollen diagram was fluvially transported from further away, the cultural character of the spectrum can be magnified as a consequence of an extensive settlement located about 5 km upstream from Libice which had existed in the 6th century (Princová 2004). In our data, it is impossible to detect the landscape change which was caused only by the activities around the stronghold at Libice. The local landscape change could either have been limited, as at the stronghold at Stará Boleslav, or it could have been more extensive which is likely due to the bigger expeted population at Libice, and especially because there was already a settlement there before the stronghold and thus human activity had an earlier stert at that site.

Hradišťko

The fact that the stronghold of Hradišťko has a longer and better sedimentary record and is situated in the same region as Libice (Fig. 1) compensates for the lack of a sedimentary record from the High Middle Ages at Libice.

Prior to the foundation of the stronghold, there was a combination of woodland and open cultural landscape around this site. Exploitation of the landscape was not intensive but rather cumulative, which is reflected by a long-term gradual reduction of woodlands, in which the taxa composition was practically undisturbed (Fig. 9). The foundation of the stronghold is reflected only by increased numbers of macrofossils of some ruderal taxa and a reduction in macrofossils of Alnus and Quercus. The rather limited local impact of the stronghold is evident from the lack of apparent change in the pollen diagram. The radical landscape transformation can be connected only with colonization of this area in the High Middle Ages which is dated to the late thirteenth or 14th century (Fig. 9), and when accumulation of flood loams in the oxbow also started. The very limited impact of the foundation of the small stronghold of Hradišťko shown by pollen data can serve as a rather interesting precedent for other studies, demonstrating that especially smaller settlements in prehistoric landscapes can be invisible when traced only by pollen data.

The results obtained from Hradišťko show that the landscape changes, which occurried in the tenth and 11th centuries and can be linked to the foundation of the nearby centre of Libice, did not affect the natural environment around Hradišťko, which was only 7.5 km away. Another settlement situated at the site of the modern city of Kolín, only 3 km upstream from Hradišťko (Fig. 1), did not profoundly affect its natural environment either.

109 These results suggest that the early medieval landscape in central Bohemia was still relatively well wooded. Individual settlements were isolated by the woods from other ones, only a few kilometres distant, even when some of the settlements were important centres. We do not know if the latter is valid for the entire old settlement zone of the modern Czech Republic, but it is certainly true for the alluvial landscapes in this study. The Czech Republic covers an area which lay outside the Roman Empire, and the history of this area, including cultural landscape development during the Roman and Migration periods, is therefore very different to that of the Roman Empire itself. One of the main differences was certainly the rate of removal of woodland and the timing of the extensive opening of the landscape, which occurred earlier in parts of western Europe which had been in the Roman Empire (Nakagawa et al. 2000; Mäckel et al. 2009; Doyen et al. 2013; Etienne et al. 2013), than in the 13th century in the area of the Czech Republic.

Medieval landscape transformation

In the Early Middle Ages, forests around the strongholds were species rich lowland mixed oak woods combined with pines on sandy soils. There were also similar forests were, according to pollen analyses, around important Moravian centres from the 9th century, such as Mikulčice and Pohansko (Svobodová 1990; Doláková et al. 2010). The woodlands in the early medieval vegetation were probably stable for as long as 3,000 years in their main characteristics and especially taxon composition. This idea is corroborated by the pollen diagram from Stará Boleslav 2 (Fig. 2; Břízová 1999) as well as by pollen data from other Czech sites located in lowlands where the pollen record from the agricultural period of the Holocene is complete (Pokorný 2005).

Macrofossil data from sites studied here have enabled a detailed reconstruction of non- woodland vegetation communities. The variability of macrofossil assemblages, especially the number of ruderal, weed and dry land taxa, was exceptional due to the proximity of populated areas to the sampled sites and there was maybe some long-distance fluvial transport too. Floral richness as well as the resulting plant communities surrounding the sites compares well with the macrofossil analyses from the stronghold of Mikulčice (Opravil 1978).

In our work we have focused on an extremely detailed study of landscape development in the course of several centuries. One of the major limits of such an effort is that radiocarbon dates are less accurate than historically or archaeologically based chronologies. This may be the reason why, at least according to our knowledge, other similar studies have not gone into such detail (Latałowa 1992; Niewiarowski 1995; Miller et al. 1997; Risberg et al. 2002). Comparisons with these studies seem to indicate that the human impact caused by the early existence of our three archaeological sites is more similar to the impact of, for instance, the Iron Age stronghold of Biskupin in Poland (Niewiarowski 1995) rather than the impact of some Early Medieval trading centres such as Wolin and Birka (Latałowa 1992; Miller et al. 1997; Risberg et al. 2002). Such comparisons can be only made with caution because the other pollen sites have a more regional pollen signal, mostly from lakes or bigger mires and there are greater distances between the pollen sites and settled areas. Also, the results sometimes derive from

110 cultural layers (Risberg et al. 2002) or they do not have radiocarbon dates (Niewiarowski 1995; Karlsson 1997).

Landscape changes that occurred approximately in the 13th century were radical and strongly influenced the pollen data from both Hradišťko and Stará Boleslav, but the sedimentary record from Libice ended in the Early Middle Ages. In the case of Stará Boleslav, the earliest significant landscape transformation had occurred as early as the 11th century, and as far as pollen data are concerned, it seems that the main colonization wave also came earlier, in the late 12th century (Břízová 1999). The major difference characterizing the High Middle Ages was, very probably, the increase in population. However, demographic data remain uncertain and thus only the consequences of massive colonization can be better observed. At least in western and central Europe, the high medieval transformation brought not only vast clearance of woodland but also irreversible changes in woodland taxon composition, with several general changes such as the expansion of Pinus, occurrence of Carpinus in places where Quercus used to be more common, reduction in some demanding deciduous trees such as Tilia and Ulmus and recent forestry plantations of Pinus and Picea (Nožička 1957; Rösch 2000; Ralska- Jasiewiczowa et al. 2004; Brown and Pluskowski 2011; Wieckowska et al. 2012).

The rapid increase in soil erosion that occurred as a consequence of massive removal of woodland most significantly changed the morphology of lowland alluvia by causing sedimentation of flood loams, an event that has been noticed in all studied profiles. Even though the summarized data from Germany also show that there were increases in soil erosion during the prehistoric period, especially in the late Bronze and Iron Ages (Dreibrodt et al. 2010), they also agree with our results in that the most intensive erosion occurred only in high medieval and modern times (Böse and Brande 2010; Dreibrodt et al. 2010).

Conclusions

Relevant literature references indicate that an increased attention should be paid to the interpretation of pollen and macrofossil data from alluvial sedimentary basins. Long-distance fluvial pollen transport may reduce the interpretation potential of such sites in relation to local events, and thus the events that are relevant for comparison with local archaeological data. By taking these fluvial effects into consideration, we have still found our pollen and macrofossil records well comparable with local archaeological data, expecially due to the fact that our sites in old oxbows were separated from the main river. At two of our three sites, the crucial early medieval part of the profile was represented by organic sediments with only a minor amount of fluvial silt.

Generally speaking, formation of a cultural landscape seems to have been limited to areas only a few kilometres wide on the alluvial land around the most densely populated strongholds in the early medieval period. According to our data, the foundation of the three strongholds could have had various impacts on the local natural environment. The site of Stará Boleslav was founded in an area of wooded alluvium and the pollen data show two distinct colonization phases. A relatively small human impact st a landscape level was detected in the 10th and the first half of the 11th centuries with a considerable increase in the later 11th century. It was unfortunately not possible to detect the impact of the presumably most densely populated

111 stronghold at Libice, with 600–900 inhabitants. due to the extreme change in sedimentation which made the comparison of pollen spectra before and after the foundation of the stronghold impossible. In any case, the amount of human activity around Libice must have been spatially limited, since it did not affect areas further than approximately 4-5 km from the site. The stronghold of Hradišťko was smaller than the other two centres with about 90–150 inhabitants, and its foundation is reflected only by a slight increase in macrofossils of a few ruderal plants. Otherwise, the pollen diagram from Hradišťko showed a gradual reduction in woodlands during the early medieval period.

Our results have clearly showen that the landscape transformation occurring on the onset of High Middle Ages was a very radical and rapid process, at least in this particular alluvial setting of central Bohemia.

Acknowledgements The research was supported by the Czech Academy of Sciences (project No.GA404/08/1696). We thank Jan Havrda for geological description of studied sediments. We are grateful to Bruce Albert and Dagmar Dreslerová for critical comments on the manuscript and to Petra Maříková Vlčková and James Greig for revising English.

References

Beug H-J (2004) Leitfaden der Pollenbestimmung für Mitteleuropa und angrenzende Gebiete. Verlag, München

Boháčová I (2001) Pražský hrad a jeho nejstarší fortifikační systémy – Die Prager Burg und ihre ältesten Befestigungssysteme. In: Ježek M, Klápště J (eds) Mediaevalia archaeologica 3, pp 179-301

Boháčová I (ed) (2003) Stará Boleslav. Přemyslovský hrad v raném středověku [Stará Boleslav. Přemyslid stronghold in the Early Middle Ages], Mediaevalia archaeologica 5. Praha

Boháčová I (2006) Stará Boleslav. Stav a perspektivy studia funkcí a prostorového uspořádání přemyslovského hradu [Stará Boleslav – stage and prospects for the study of the function and spatial organisation of the Přemyslid stronghold], Archeologické rozhledy 58, 695-723

Boháčová I (2008) Contribution to the study of hinterland of the Early Medieval Stará Boleslav. In: Poláček L (ed) Das wirtschaftliche Hinterland der frűhmittelalterlichen Zentren, Internationale Tagungen in Mikulčice 6, pp 179-194

Boháčová I (2010) Die Kirchen auf dem Burgwall von Stará Boleslav. Ein Beitrag der Archäologie zur Erforschung der frühmittelalterlichen Sakralarchitektur in Böhmen. In: Poláček L, Maříková-Kubková J (eds) Frühmittelalterliche Kirchen als archäologische und historische Quelle, Spisy Archeologického ústavu AV ČR Brno, Internationale Tagungen in Mikulčice 8, pp 243-261

Boháčová I (2011) Prague, Budeč and Boleslav. The reflection of state formation in Early Medieval archaeological sources. In: Macháček J, Ungerman Š (eds) Frühgeschichtliche Zentralorte in Mitteleuropa, Studien zur Archäologie Europas 14, Bonn, pp 371-395

112 Boháčová I, Hošek J (2009) Raně středověké nože ze Staré Boleslavi, Frühmittelalterliche Messer aus Stará Boleslav (Altbunzlau). Archaeologia historica 40, 367-392

Bonny AP (1978) The effect of pollen recruitment processes on pollen distribution over sediment surface of a small lake in Cumbria. J Ecol 66: 385-416

Böse M, Brande A (2010) Landscape history and man-induced landscape changes in the young morainic area of the North European Plain – a case study from the Bäke Halley, Berlin. Geomorphology 122: 274-282

Brown AG, Carpenter RG, Walling DE (2007) Monitoring fluvial pollen transport, its relationship to catchment vegetation and implications for palaeoenvironmental studies. Rev Palaeobot Palynol 147: 60-76

Brown A, Pluskowski A (2011) Detecting the environmental impact of the Baltic Crusades on a late-medieval (13th-15th century) frontier landscape: palynological analysis from Malbork Castle and hinterland, Northern Poland. J Archaeol Sci 38: 1957-1966

Břízová E (1999) Late Glacial and Holocene development of vegetation in the Labe river floodplain (Central Bohemia, Czech Republic). Acta Palaeobotanica 1999, Suppl 2: 5-12

Čulíková V (1999) Rostlinné makrozbytky z objektu 126 na předhradí slovanského hradiska v Libici nad Cidlinou [Plant macrofossils in the object no. 126 in Libice nad Cidlinou]. Památky archeologické 90: 166-185

Čulíková (2003) Rostlinné makrozbytky z raně středověkého hradu Stará Boleslav [Plant macrofossils from Stará Boleslav]. In: Boháčová I (ed) Stará Boleslav, Přemyslovský hrad v raném středověku [Stará Boleslav, Přemyslid stronghold in the Early Medieval Age]. Medievalia Archaeologica 5, pp 367-379

Čulíková V (2006) Rostlinné makrozbytky z prostoru raně středověkého opevnění v sondě 236 na jz. okraji předhradí v Libici nad Cidlinou [Plant macrofossils from the Early Medieval fortification of the Libice nad Cidlinou]. Archeologické rozhledy 58: 527-539

Danzeglocke U, Jöris O, Weninger B (2012) CalPal-2007online. http://www.calpal-online.de/, accessed 2012-05-18

Doláková N, Roszková A, Přichystal A (2010) Palynology and natural environment in the Pannonian to Holocene sediments of the Early Medieval centre Pohansko near Břeclav (Czech Republic). J Archaeol Sci 37: 2538-2550

Dreibrodt S, Lubos C, Terhorst B, Damm B, Bork HR (2010) Historical soil erosion by water in Germany: Scales and archives, chronology, research perspectives. Quatern Int 222: 80–95

Dreslerová D, Břízová E, Růžičková E, Zeman A (2004) Holocene environmental processes and alluvial archaeology in the middle Labe (Elbe) valley. In: Gojda M (ed) Ancient Landscape,

113 Settlement Dynamics and Non-Destructive Archaeology. Czech research Project 1997-2002. Academia, Praha, pp 121-171

Faegri K, Iversen J (1989) Textbook of pollen analysis. J. Wiley & Sons, Chichester

Firbas F (1949) Spät- und nacheiszeitliche Waldgeschichte Mitteleuropas nördlich der Alpen. I. Allgemeine Waldgeschichte. Jena

Giesecke T, Bennett KD, Birks HJB, Bjune AE, Bozilova E, Feurdean A, Finsinger W, Froyd C, Pokorný P, Rösch M, Seppä H, Tonkov S, Valsecchi V, Wolters S (2011) The pace of Holocene vegetation change – testing for synchronous developments. Quaternary Sci Rev 30: 2805-2814

Grimm EC (1992) Tilia and Tilia-graph: pollen spreadsheet and graphic programs. Program and Abstracts, 8th International Palynological Congres, Aix-en-Provence, France, September 6-12, p 56

Hassan FA (1978) Demographic Archaeology. In: Advances in Archaeological Method and Theory, Vol. 1. Academic Press, Inc., New York, pp 49-103

Karlsson S (1997) Pollen analysis from a rock depression, the hillfort, Birka, Björkö. In: Miller U, Clarke H, Hansson A-M, Johansson BM, Ambrosiani B (eds) Environment and Vikings: scientific methods and techniques. PACT 52/Birka Studies 4, pp 239-248

Klápště J (2012) The Czech Lands in Medieval Transformation. Brill, Leiden – Boston

Kolb F (2005) Die Stadt im Altertum, Albatros Verlag, Düsseldorf

Kuna M, Profantová N (eds) (2005) Počátky raného středověku v Čechách [Beginning of Early Middle Ages in Bohemia]. Archeologický ústav AV ČR, Praha

Latałowa M (1992) Man and vegetation in the pollen diagrams from Wolin Island (NW Poland). Acta Palaeobot 32 (1): 123-249

Macháček J (2005) Raně středověké Pohansko u Břeclavi – munitio, palatium nebo emporium moravských panovníků? [Early Medieval centre in Pohansko near Břeclav/Lundenburg: munitio, paladium or emporium of the rulers of Moravia?]. Archeologické Rozhledy 57: 100- 138

Macháček J, Doláková N, Dresler P, Havlíček P, Hladilová Š, Přichystal A, Roszková A, Smolíková L (2007) Raně středověké centrum na Pohansku u Břeclavi a jeho přírodní prostředí [Early Medieval centre at Pohansko near Břeclav and its natural environment]. Archeologické rozhledy 59 (2): 278-314

114 Mařík J (2009) Libická sídelní aglomerace a její zázemí v raném středověku. Early Medieval Agglomeration of Libice and its Hinterland. Dissertationes Archaeologicae Brunenses/ Pragensesque

Mařík J (2013) From central Places to power domain, development of Early Mediavel landscape on middle Elbe and lower Cidlina. Zentrale Orte und zentrale Räume des Frühmittelalters, in press.

Mlíkovský J (2003) Zvířata a jejich role na raně středověkém hradě Stará Boleslav, střední Čechy [Animals and their role at the stronghold of Stará Boleslav]. In: Boháčová I (ed) Stará Boleslav [Stará Boleslav. Přemyslid stronghold in the Early Middle Ages]. Přemyslovský hrad v raném středověku. Medievalia archaeologica 5: 347-365

Miller U, Clarke H, Hansson A-M, Johansson BM, Ambrosiani B (1997) Environment and Vikings: scientific methods and techniques. PACT 52/Birka Studies 4, 275 pp

Niewiarowski W (ed) (1995) Zarys zmian środowiska geograficznego okolic Biskupina pod wpływem czynników naturalnych i antropogenicznych w późnym glacjale i holocenie [Outline of changes of the geographical environment in the Biskupin surroundings under influence of natural and anthropogenic factors during the Lateglacial and Holocene]. Oficyna Wydawnicza Turpress, Toruń

Nožička J (1957) Přehled vývoje našich lesů [Development of forests in the Czech Republic]. Státní zemědělské nakladatelství, Praha

Opravil E (1978) Rostlinná společenstva v okolí Mikulčic v období předvelkomoravském a velkomoravském [Plant communities in the vicinity of Mikulčice before and during the Great Moravian Empire]. Archeologické rozhledy 30: 67-75

Opravil E (1983) Údolní niva v době hradištní [Alluvium during the Early Middle Ages]. Studie Archeologického ústavu Československé akademie věd v Brně. Academia, Praha

Peck RM (1973) Pollen budget studies in a small Yorkshire catchment. In: Birks HJB, West RG (eds) Quaternary Plant Ecology. Blackwell Scientific Publications, Oxford, pp 43-60

Pennington W (1979) The origin of pollen in lake sediments: an enclosed lake compared with one receiving inflow streams. New Phytol 83: 189-213

Pokorný P (2005) Role of man in the development of Holocene vegetation in Central Bohemia. Preslia 77: 113-128

Princová J (2004) Sídliště ze 7. až počátku 9. století v poloze „Staré Badry“ u Opolánek, okr. Nymburk. Výsledky výzkumu v letech 1965-1966 a 1969-1973 – Siedlung aus dem 7. bis Anfang 9. Jahrhundert in der Flur „Staré Badry“ bei Opolánky, Bez. Nymburk. Ergebnisse der Ausgrabungen in den Jahren 1965-1966 und 1969-1973. Památky archeologické 95: 107-174

115 Punt W (ed) (1980) The Northwest European Pollen Flora, Vol 37, Umbelliferae. Elsevier, Amsterdam

Ralska-Jasiewiczova M, Latałowa M, Wasylikowa K, Tobolski K, Madeyska E, Wright HE, Turner C (eds) (2004) Late Glacial and Holocene history of vegetation in Poland based on isopollen maps. Krakow

Reille M (1992) Pollen et spores d’Europe et d’Afrique du nord. Marseille

Risberg J, Karlsson S, Hansson A-M, Hedenström A, Heimdahl J, Miller U, Tingvall C

(2002) Environmental changes and human impact as recorded in a sediment sequence offshore from a Viking Age town, Birka, southeastern Sweden. The Holocene 12 (4): 445-458

Rösch M (2000) Long-term human impact as registered in an upland pollen profile from southern Black Forest, south-western Germany. Veget Hist Archaeobot 9: 205-218

Sádlo J, Pokorný P, Hájek M, Dreslerová D, Cílek V (2008) Krajina a revoluce [Landscape and Revolution]. Malá Skála, Praha

Sláma J, (1988) Střední Čechy v raném středověku. Archeologie o počátcích přemyslovského státu [Central Bohemia during the Early Middle Ages. Archaeology of the early Czech state], Praehistorica 14, Praha

Steuer H (1984) Zur ethnischen Gliederung der Bevölkerung von Haithabu anhand der Gräberfelder, Offa 41: 189-209

Stloukal M, Vyhnánek L (1976) Slované Velkomoravských Mikulčic [Slavs from the Great Moravian Mikulčice], Academia, Praha.

Svobodová H (1990) Vegetace jižní Moravy mezi 500-1000 AD [Vegetation of the south Moravia betweed AD 500-1000]. Archeologické Rozhledy 42: 170-205

Sugita S (1994) Pollen representation of vegetation in Quaternary sediments: Theory and Method in patchy vegetation. J Ecol 82: 881-897

Tolonen K (1986) Charred particle analysis. In: Berglund BE (ed) Handbook of Holocene palaeoecology and palaeohydrology. John Wiley & Sons, Chichester, pp 485-496 Wieckowska M, Dörfler W, Kirleis W (2012) Vegetation and settlement history of the past 9000 years as recorded by lake deposits from Großer Eutiner See (Northern Germany). Rev Palaeobot Palynol 174: 79-90

Paper II

116 The oldest Czech fishpond discovered? An interdisciplinary approach to reconstruction of local vegetation in medieval Prague suburbs

Adéla Pokorná, Petra Houfková, Jan Novák, Tomáš Bešta, Lenka Kovačiková, Kateřina Nováková, Jan Zavřel, Petr Starec

117 The oldest Czech fishpond discovered? An interdisciplinary approach to reconstruction of local vegetation in medieval Prague suburbs

Adéla Pokorná, Petra Houfková, Jan Novák, Tomáš Bešta, Lenka Kovačiková, Kateřina Nováková, Jan Zavřel, Petr Starec

Abstract

Wet sediments of a former water reservoir were discovered during an archaeological rescue excavation. Vegetation and environmental changes taking place in the medieval suburbs of Prague, Czech Republic, from the 10th to the middle of the 14th century were investigated. The origin and function of the water reservoir was revealed using a multi-proxy approach that combined the results of macrofossil, pollen, diatom, antracological, archaeolo-zoological and sedimentological analyses.

Gradual changes of the surrounding vegetation were documented. Field indicators increased in time, whereas proportions of broad-leaf trees and shrubs decreased; proportions of ruderal plants increased continually. A gradual decline of semi-natural hygrophilous vegetation was accompanied by an inverse tendency in trampled vegetation. All these trends indicate an intensification of human activity around the pool.

A similar intensification of anthropogenic influence is clearly visible in the development of the aquatic environment. According to the diatom composition, the base of the profile is the result of sedimentation in considerably oligotrophic conditions. A successive deterioration of water quality was documented by various organisms (diatoms, green algae, water macrophyta, fishes, and intestinal parasites). The high content of dissolved nutrients, probably connected with anoxia, could have caused the disappearance of both diatoms and fishes.

Introduction

Fish farming has a long and rich tradition in the Czech lands. The spreading of carp (and knowledge of carp breeding) was probably connected with cultural influences coming from West Europe (Makowiecki 2001). The progressive development of fishpond building in 13th century further culminated in 15th to 16th centuries (Andreska 1987, Čítek et al. 1998, Hurt 1960). However, there are indications that some kind of artificial water reservoirs containing fishes existed here already before the introduction of the carp-fishponds. Sporadic written sources (mostly chronicles and legends) mention “fishponds” already in 10th to 11th centuries. The oldest known reference is traditionally being identified with a ceased Medieval village called Rybník (the Czech meaning of its name is “the Fishpond”) once located in the area of today’s New Town of Prague. The Rybník village was first mentioned in 993 in the Donation Deed of the Břevnov Monastery, obliging that the monastery receives tithes from all fields belonging to this village (called Ribnyk in the original Latin text).

A range of palaeoenvironmental techniques was used to study sediments which were supposed to be remnants of a pool once connected with the Rybník village. Among other,

118 diatom analysis and the presence of eggs of intestinal parasites were included in the presented research. Structure of diatom community is tightly linked with changes in aquatic environment (e.g. Battarbee 1986) and the rate of artificial pollution has been traditionally expressed in the index of saprobity (Sládeček 1986). The presence of eggs of intestinal parasites has been used as an indicator of pollution by feces (Bosi 2011).

Combination of several techniques, above all palynology, anthracology and analysis of plant macro remains was used for reconstruction of both aquatic and terrestrial vegetation. These analyses already prooved their reliability within the palaeoecological and archaeobotanical research (e.g. Smol et al. 2001). Each of these methods naturally has its own advantages as well as limits resulting from different source area of material, taphonomy and different determination level (for more details see e.g. Birks and Birks 2006, Gaillard 2007, Novák et al. 2012, Théry-Parisot et al. 2010), therefore it is advisable to combine them (for some examples see Sadori et al. 2010 or Święta-Musznicka et al. 2013).

Good examples of former water reservoirs studied using multi-proxy techniques are investigations of late-Glacial to early-Holocene Lake Švarcenberk (Pokorný 2002) or a cistern at the pre-Roman Iron Age hillfort Vladař (Pokorný et al. 2006) as well as a Holocene profile from the Řežabinec fishpond (Rybníčková and Rybníček, 1985). However, there are not many palaeoenvironmental studies dealing with Medieval fishponds. Several fishponds were investigated in Germany (Hellwing 1997, Rösch 1999, 2012). Sediments of Medieval fishpond Vajgar (Jindřichův Hradec, Czech Republic), established at the 13th century, reflect Medieval colonization as well as the development of the fishpond until recent times (Jankovská and Pokorný 2002).

The study site

The foundation of the New Town of Prague in 1348 had probably posed prominent changes in the environment of the Medieval town’s suburbs. According to its Foundation Deed, the New Town was established in a suburban area, where “villages, gardens and fields had been located”. We attempt to examine what the vegetation character of these suburbs must have been like before the site changed to one of building and construction.

The New Town of Prague (as well as the Old Town of Prague) is situated on the right bank of the Vltava River (Fig. 1). The morphology and character of its original terrain, as well as its hydrology, was directly influenced by a system of several riverine terraces, deposited here within the Pleistocene. According to many archaeological records, several moist depressions and a rather dense network of brooks were evidenced in the area outside the Old Town walls (Kaštovský et al. 1999, Starec 2005, Starec et al. 2012, Kašpar 2007). The moist depressions were later filled in with the town’s domestic waste and likely disappeared even before the foundation of the New Town (Hrdlička 1984, 1997). The permanently moist sediments of these features represent an optimal environment for the preservation of both plant macrofossils and pollen grains. However, the waste origin of these deposits makes the interpretation of the performed archaeobotanical analyses difficult, as the plant assemblages represent an inseparable mixture of material from several sources.

119 Having the intention to reconstruct the character of the area’s original vegetation, we needed to analyse a different type of sediment, namely the sediment of a water reservoir that had resulted from a natural sedimentation process. In 2009, such sediment was found during a rescue archaeological excavation in the cellar of the house V Tůních no. 1625/II (the excavation being directed by P. Starec, Prague City Museum). The formation of grey clay layers was preliminarily interpreted as the lacustrine sediments of a former pool, promising good preservation of bio-facts in the waterlogged conditions. These sediments (and the pool) could have been related to the ceased Medieval village Rybník (see above). Archaeological evidence (Kašpar 2003) supports the existence of the village at a location to the northwest of the investigation site between at least the 11th and 13th Century (Fig. 1, 2).

Our aims are to describe the vegetation of the Medieval Old Town of Prague’s suburbs and to consider the environmental changes that have occurred before the mid-14th century. The second aim of this study is to reveal the origin and function of the investigated water reservoir.

Materials and methods

A trench 1 m wide and approximately 1 m deep had been dug out during the reconstruction of a basement (canalisation trench) of the house V Tůních no. 1625/II. Four profiles, placed subsequently on the line of the trench, and one additional profile placed within the static test pit, were studied (for the position of individual profiles, see Fig. 2. As all the encountered sediments were waterlogged, sediment samples taken from individual layers were wet sieved to obtain biological remains (seeds, charcoal pieces, animal bones, etc.) Two sieves of 1 mm and 0.25 mm were used, sample volumes varying from 0.5 to 3 l. Fragments of wood and animal remains (bones of large and medium-sized mammals, mussels) when encountered during the excavation were also analysed.

For a more detailed approach, a box profile one meter in length was taken within the profile A20 (the longest of the profiles investigated). Sub-samples of 1 cm3 were taken for pollen and diatom analyses at 5 cm intervals. In the lowermost part of the profile, the sampling frequency was increased. The remaining sediment was then divided into 5 cm sections (approx. 0.5 l each) and wet sieved to obtain plant macrofossils and fish remains. Three sub-samples for radiocarbon dating were also taken from this profile. Radiocarbon (14C AMS) dating was undertaken using selected Chenopodium album seeds by the laboratory CAIS, USA (Center for Applied Isotope Studies, University of Georgia). Calibration of radiocarbon data, based on the calibration curve IntCal 09 (Reimer et al. 2009), was performed using the application OxCal 4.1.6 (Ramsey 2010).

120

Fig 1 Schematic representation of the surroundings of the Old town of Prague before 1348. The boundary of the New town of Prague is indicated by dotted line. The former settlements’ positions are marked by dashed lines. The supposed position of Rybnik village is marked by an arrow. The course of the Vltava River (grey) corresponds to the recent situation. Redrawn according to Mencl (1969)

121

The wet-sieved and dried material was sorted under a binocular microscope and the plant seed/fruit remains were identified in the whole volume, using both a reference collection (Department of Botany, Faculty of Science, Charles University in Prague) and determination literature (Cappers et al. 2006, Katz et al. 1965). The results of the carpological analysis were processed using the ArboDatMulti database program (Kreuz and Schäfer 2002) and became a part of the Czech archaeobotanical database CZAD (Pokorná et al. 2011). The nomenclature used in the following text was based on Kubát 2002 (vascular plants) and Chytrý and Tichý 2003 (vegetation units).

Fig 2 Surroundings of the excavation site including positions of individual profiles mentioned in the text. The Romanesque rotunda of Saint Stephen (Saint Longinus today) represented the religious centre of the Rybnik settlement. A section of a former stream (grey) was documented archaeologically by Kašpar (2007). The supposed direction of the stream (according to the present-day slope direction) is shown by an arrow Pollen grains from sub-samples (1 g) were extracted by chemical treatment according to Faegri and Iversen (1989) including boiling in 10% KOH, sieving, acetolysis and treatment in 40% HF to remove silica. Lycopodium spores of a known quantity were added to each sample in order to determine the absolute pollen concentration (Stockmarr 1971). Pollen grains were counted under a light microscope at a magnification of 400-1000x. A minimum of 500 pollen determinations (where possible) were made for every sample. Taxonomic identifications followed Punt (1976), Punt and Blackmore (1991), Punt and Clarke (1980, 1981, 1984), Punt et al. (1988, 1995, 2003, 2009), and Beug (2004). The presence of non-pollen palynomorphs (e.g. intestinal parasite eggs, algal and invertebrate remains) was recorded (Komárek and Jankovská 2001, van Geel 2001); its quantification was related to the total pollen count. Data were processed using Tilia 1.5.12. software (Grimm 2011). The diversity of pollen spectra was expressed using the Shannon index (Shannon 1948).

The charcoal and wood analyses were performed only on the largest fraction of fragments (>2 mm). Charcoal pieces were identified using an episcopic interference microscope (Nikon Eclipse 80i) with a 200-500x magnification. The reference collection (Laboratory of Archaeobotany and Paleoecology in České Budějovice) was used for determination. In addition, standard identification keys were also used (Schweingruber 1990, Heiss 2000).

140 The analysis of diatoms was based on methods described by Battarbee (1986). Permanent slides were prepared using hydrogen peroxide for digestion, and Pleurax as a mounting medium (Fott 1954). Frustules were counted using a light microscope at a magnification of 1000x. A minimum of 400 valves were counted per slide with the exception of the deepest sample where the total only reached 99 valves due to the low concentration of frustules in the sediment. Taxonomic identifications primarily followed Süsswasserflora von Mitteleuropa (Krammer and Lange-Bertalot 1986, 1988, 1991a, 1991b). Calculation of saprobic indices was based on values of species specific weights and saprobic indices stated in Sládeček (1986) with corrections proposed by Marvan (unpublished).

Small fish remains (predominantly scales, pharyngeal teeth and vertebrae) recovered by wet-sieving were identified using a fish skeleton reference collection (Laboratory of Archaeobotany and Paleoecology in České Budějovice) and determination literature (Baruš et al. 1995, Radu 2005, Wheeler and Jones 2009). Fish remains were identified to the lowest taxonomic level possible. The evaluation of representative taxa was calculated using the number of identified specimens (NISP). For the determination of aquatic bivalve molluscs, Pfleger 1988 and Beran 1998 were used. An overview of the identification of mammalian bones is summarized elsewhere (Kovačiková 2009, unpublished report).

A CONISS cluster analysis (Grimm 1987) and determination of significant pollen accumulation zones (PAZ) based on the broken stick model (Mac-Arthur 1957, Legendre & Legendre 1998) was performed using package Rioja (Juggins 2009) in R 2.11.0 (R Development Core Team 2010). Canonical correspondence analysis (CCA) of pollen data was realized in the package Vegan (Oksanen et al. 2010). Affiliation of the samples to particular sediment layers was used as an environmental factor. Both cluster and CCA analyses were performed on square root transformed data.

Fig 3 Illustration of the four profiles mentioned in the text (for detailed position of the profiles see Fig. 2), with representation of organic remains indicating the aquatic phase of sedimentation. The position of sampling boxes is represented by the two dashed line rectangles in the profile A20

Table 1 General description of individual zones distinguished within the profile A20

Zone Mineralogical description Eco-facts presence

141 1 Sandy layers with bulky quartz and Very low concentrations of plant macro remains and quartzite fluvial boulders no fish remains Frequent presence of various indicators of water Grey dusty clays with high phase: seeds of various species of aquatic plants admixture of organic material, 2 (Zannichellia pallustris, Potamogeton spp., classified by its character and Ceratophyllum demersum, Lemna minor / gibba), fish stratigraphic position as sediments remains, mussels, Cladocera, Charales, coccal green of a pool with standing water algae, and diatoms A thick formation consisting of sand No fish remains, seeds of Lemna minor / gibba in 3 with stripes of grey dusty clay in the much lesser quantity than in the zone 2, no other

upper part of the profiles aquatic plants

Results

General features

Three main vertically positioned zones were distinguished within the whole profile according to their sedimentary structure and the presence or absence of eco-facts indicating the aquatic environment (Fig. 3, Table 1). Zone 1 overlies the bedrock formed by the Older Palaeozoic marine sediments of the Prague Basin (Barrandien). The gradual decrease of its thickness towards the northwest implies that it could represent deluvio-fluvial sediments transported to the site from the Vinohrady terrace (the edge of this riverine terrace of Pleistocene origin is about 150 m to the southeast). Zone 2 surely represents the aquatic phase of the sedimentation (for more details see below). Zone 3 delineates a well-marked sedimentary stratification, probably the result of an outwash erosion of psammitic material and its subsequent deposition in the water reservoir.

Table 2 The environmental demands of fishes and other aquatic animals documented archaeo-zoologically

Taxon Characteristics

Fishes

Rutilus rutilus Prefers deeper water layers; young individuals stay in the shallows; the middle roach - Cyprinidae tolerance of oxygen dissolved in water; maximum length 40 cm

Prefers warmer parts in shallow shore zone; able to tolerate low oxygen Tinca tinca concentrations; present in both stagnant and slow-moving water, often in pools tench - Cyprinidae with muddy substrates, covered with abundant vegetation; maximum length 30-63 cm

Prefers indented bottoms and banks (it hides both between stones and under Leuciscus cephalus the undermined and overgrown shores); it tolerates slightly polluted water, if chub - Cyprinidae the amount of dissolved oxygen is sufficient; maximum length 60 cm

Alburnus alburnus Occurs near the surface (superficial fish) of both stagnant and slow-moving bleak - Cyprinidae water, avoiding the parts overgrown by vegetation; maximum length 15-20 cm

Three species could be expected within this family: Perca fluviatilis (European perch) - lakes of all types to medium-sized streams; Sander lucioperca (pike- Perciformes perch) - large, turbid rivers and eutrophic lakes; and Gymnocephalus cernuus (ruffe) - bottom parts of both lacustrine and flowing waters (depths to 85 m)

142 Mussels

Anodonta/Unio Mussels of family Unionidae occur mostly in slow-mowing watercourses of mussel - Unionidae different types as well as in stagnant water

Cladocera

Typical zooplankton of various types of stagnant water; it tolerates highly Daphnia magna polluted water; often being found in enormous amounts (up to 2000 individuals/L)

Daphnia pulex In slightly polluted ponds and backwaters

Within the archaeozoological assemblage, 82.6 % (NISP=166) of the total remains (N=201) have been identified as fish (the remains comprising bones, teeth and scales). Scales (both cycloid and ctenoid) and their fragments dominated (55.2 % of total finds of fish remains). The identified species included tench (Tinca tinca), chub (Leuciscus cephalus), bleak (Alburnus alburnus) and roach (Rutilus rutilus). The ctenoid scales are typical of the order Perciformes. From this order, we take into consideration the following species: European perch (Perca fluviatilis), pike-perch (Sander lucioperca) or ruffe (Gymnocephalus cernuus). For the environmental demands of the fishes (and other aquatic animals identified) see Table 2.

Profile A20

The profile A20 was used for more detailed investigation because it was the longest one, and also because the zone 2 was the thickest one in this profile. For precise position of the two sampling boxes see Fig. 3. Three samples were chosen for dating using the radiocarbon method (see Table 3). The seeds of Chenopodium album were used in all cases. The lowermost layer of the profile was not suitable for dating because of a very low concentration of plant macro remains (and absence of the terrestrial ones). Therefore, the first sample was taken from the layer 180 (10 cm above the bottom). The second sample was taken from the the layer 178 (40 cm above the bottom). Those two dates delimit the duration of the sedimentation of the zone 2 between the end of 10th century and the fist half of 14th century. The third sample for dating was located 65cm above the bottom (the zone 3, layer 161). This sample was chosen to determine the period of sedimentation of the thin clayey layer between the sandy layers overlaying the aquatic sediments.

Two significant pollen accumulation zones (PAZ) were determined using species with minimum abundance of 3%. On the contrary, no significant PAZ occurred if all species were included. The classification of A20 profile according to the sedimentologicaly defined layers was preferred, since the results of CCA analysis showed their significant (p<0.05, 39.1% of total explained variability) relation to the pollen samples composition.

Layer 182

143 The lowermost layer was characterized by an absence of pollen grains and a very low occurrence of plant macro-remains. However, Lemna minor/gibba, Zannichellia palustris and several pieces of Salix branches were found here.

Layer 181

The concentration of plant macro-remains was still very low here; the pollen spectrum was composed mainly of non-arboreal pollen (NAP). Following macrophytes were documented (both macro-remains and pollen): Alisma plantago-aquatica, Lemna minor/gibba, Myriophyllum spicatum type, Potamogeton (P. natans type), Sparganium erectum type and Zannichellia palustris. The diatom composition (Amphora pediculus, Gomphonema sarcophagus, G. angustatum s.l., Cymbella aspera, Navicula radiosa s.l. and Planothidium frequentissimum) indicates clean water (the index of saprobity equates to 1.2). Among the water algae, Charales, Pediastrum simplex and Tetraedron minimum were found. The resting eggs of Rotifers appeared here in low quantities, as well as ephippia of Daphnia cf. pulex.

Among pollen of the terrestrial plants the family Poaceae dominated. Anthropogenic indicators (e.g. crops and numerous ruderal species) as well as grazing indicators (Plantago lanceolata, Calluna vulgaris) were recorded in the pollen spectrum. The arboreal/non-arboreal pollen ratio (AP/NAP) equals 25 %; charcoal of Quercus and small Salix branches were recorded.

Layer 180

The radiocarbon date from this layer has been calibrated and the resulting interval is A.D. 945 -1023 (see Table 3). The layer shows an increasing concentration of macro-remains and pollen of aquatic macrophytes (Ceratophyllum demersum, Lemna, Potamogeton), green algae (Tetreadron minimum, Pediastrum simplex and Scenedesmus) were also abundant here. Fish bones and scales were found in high quantities (e.g. Tinca tinca, Alburnus alburnus or Leuciscus cephalus) as well as mussel shells and ephippia of both Daphnia magna and D. cf. pulex. The composition of diatoms differed markedly from layer 181. Species with a low index of saprobity (Amphora pediculus and Gomphonema sarcophagus) remained only in minor quantities. A domination of Stephanodiscus hantzschii, Amphora veneta and Cocconeis placentula v. euglypta (species with an index of saprobity above 2), along with Achnanthidium minutissimum, Hippodonta capitata, Lemnicola hungarica and Nitzschia fonticola was recorded.

Among terrestrial plants, hygrophilous vegetation and wet meadows were indicated (e.g. Carex, Juncus, Lysimachia vulgaris type, Lythrum salicaria type, Persicaria hydropiper, Ranunculus sceleratus, Ranunculus acris-group, Urtica dioica type, and Scirpus sylvaticus). The abundance of pollen types of the family Poaceae and genus Artemisia decreased in contrast to the increasing percentage of anthropogenic indicators (Polygonum aviculare, Chenopodium spp., Rumex acetosa type and the family Brassicaceae). Aside from other cereal species, a

144 presence of Secale cereale pollen was recorded. Three bones of large mammals were also found here. AP/NAP remained similar to the previous layer.

Table 3 List of Calibrated Radiocarbon Data originating from macro-remains of Chenopodium album separated from profile A20 sample ID layer δ13C,‰ 14C age ± calibrated 14C age

18_A20 180 -28.2 1060 B.P. 25 cal A.D. 945 -1023

12_A20 178 -26.8 650 B.P. 25 cal A.D 1282 -1362

07_A20 161 -27.5 980 B.P. 25 cal A.D .995 - 1154

Layer 179

The layer was characterized by a high number of pollen types and a high absolute concentration of pollen grains. Also the concentration (above 500/l) of plant macro-remains and the number of determined plant species (46 taxa) reached its maximum here. Lemna dominated among aquatic macrophytes, Ceratophyllum demersum, Myriophyllum spicatum type, Potamogeton and Zannichellia palustris were abunndant too. The number of fish remains (Cyprinidae and Perciformes) also reached its maximum as well as the ephippia of both Daphnia species. Both green algae and diatoms showed a very similar species composition to that in layer 180.

Hygrophilous plant species were also abundant here (Bidens tripartita type, Eleocharis palustris aggr., Lycopus europaeus, Persicaria hydropiper, Persicaria lapathifolia, Ranunculus sceleratus, and cf. Solanum dulcamara). However, Alisma plantago-aquatica type and Sparganium erectum type decreased in the pollen spectrum. Wet meadows were represented by Chaerophyllum hirsutum type, Filipendula type, Juncus, Lychnis flos-cuculi, Peucedanum palustre type, Ranunculus acris-group, Ranunculus flammula-group, Scirpus sylvaticus, and Valeriana officinalis type.

Field crops (Horedeum type, Triticum type, Avena type, Secale cereale) and weeds as well as ruderal taxa (e.g. Artemisia type, Anthemis arvensis type, Brassicaceae, Chenopodiaceae, Polygonum aviculare type, and Rumex acetosa type) were documented in pollen. Abundance of Poaceae and Plantago lanceolata (grazing indicator) was recorded. Among the macro-remains, following weeds were found: Agrostemma githago, Anthemis cotula, Chenopodium spp., Polycnemum arvense, Thlaspi arvense, and. Valerianella dentata. Compared to layer 180, Polygonum aviculare, Chenopodium album and Urtica dioica macro- remains slightly increased. One seed of Cucumis melo/sativus and a glume of Panicum miliaceum were found here, as well as macro-remains of edible wild species: Rubus idaeus, Rubus fruticosus aggr., cf. Fragaria, Sambucus nigra and Corylus avellana.

Tree species composition in the pollen spectrum remained similar to the previous layer; however, Fraxinus type and Acer type appeared here. Besides, occurrence of rare taxa such as Prunus type and Cornus mas type was recorded. Salix pollen grains disappeared in contrast to many small pieces of Salix branches found here. This layer contained a high amount of charcoal

145 fragments of Quercus, Pinus sylvestris and Fagus sylvatica. Fragments of wood of Pinus sylvestris were quite abundant and some scarce occurrences of Abies alba wood were also recorded.

Layer 178

The radiocarbon date from this layer is cal. A.D. 1282 -1362 (see Table 3). It was still rich in both plant macro-remains and pollen types, however, the number of fish remains decreased markedly. Lemna and Zannichellia palustris remained at comparable high concentrations as in the previous layer. On the contrary, the macro-remains of Potamogeton decreased rapidly. All the diatom species disappeared in this layer, as well as the green algae (with the single exception of Pediastrum simplex). Ephippia of Daphnia cf. pulex disappeared and the D. magna concentration decreased. The occurrence of parasite eggs of Trichuris was firstly recorded from the upper half of this layer.

Macro-remains of hygrophilous plant species decreased compared to the previous layer (both the number of species and concentration). The indicators of wet meadows nearly disappeared here, also the pollen ratio of Poaceae decreased. Among the grassland species, only those of dry grasslands (Festuco-Brometea) remained present (macro-remains of Arenaria serpyllifolia and Hypericum perforatum). On the other hand, the abundance of pollen of field crops increased, as well as Polygonum aviculare type (indicator of trampled vegetation). Also the concentration of Chenopodium album increased markedly in this layer. Weeds were represented by Anthemis arvensis, Centaurea cyanus, Chenopodium ficifolium, Consolida ambigua type, Euphorbia helioscopia, Fumaria officinalis, Papaver rhoeas type, Polycnemum arvense, Thlaspi arvense and Stellaria media. One seed of cf. Vitis vinifera and two bones from a large mammal were found here.

Layer 162

Contrarilly to the previous clayey layers, this layer was characterized by a high admixture of sand. Concentration of both plant macro-remains and pollen decreased markedly here. Further, the pollen curves based on absolute values of all woody species were lowered. However, the percentage diagram shows an increase in the AP/NAP ratio. This is based on an increase in the Pinus proportion in the layers with a sand admixture.

Remains of aquatic macrophytes as well as hygrophilous plants nearly disappeared in this layer (with exception of Lemna and Ranunculus sceleratus), also Daphnia magna and rotifers became scarce here, and only two fish bones of Cyprinidae were found. On the other hand, the macro-remains of ruderal plants increased their concentration; this trend was most prominent in Chenopodium album, C. hybridum and Sambucus ebulus. The composition of root crop weeds was very similar to that of layer 178, whereas the macro-remains of field weeds were very scarce. Among the NAP pollen curves, an increase in the abundance of Chenopodiaceae, Polygonum aviculare type and Centaurea cyanus was observed.

Layer 161

146 This layer differs from both the underlying and the overlaying layers in its sedimentological character. It represents a streak of gray dusty clay between shale and siliceous sands of the layers 162, 145 and 144.

Radiocarbon date (Table 3) from this layer (after calibration) is A.D. 995 - 1154 (for more details, see discussion). In the pollen diagram, the total number of species (pollen types) decreased compared to previous layers as well as the number of plant species in the macro- remain assemblage.

The concentration of Lemna macro-remains was relatively high but no other aquatic macrophytes were encountered (except for 2 achenes of Alisma plantago-aquatica). The following species of natural hygrophilous habitats were found here: Carex, Juncus, Lemna, Persicaria maculosa, Ranunculus aquatilis-group, Ranunculus sceleratus, Rumex crispus/obtusifolius, Sparganium erectum, and Typha latifolia.

The macro-remains concentration of Chenopodium album (and other species of this genus) culminated here. Composition of the macro-remains of weeds was similar to the previous layers; on the other hand, indications of weeds such as Dipsacus fullonum type, Gnaphalium uliginosum type, Jasione montana type, Scleranthus annuus, Sedum type, Teucrium and Xanthium strumarium type appeared among the rare species in pollen spectra. The pollen of crops, weeds and ruderals as well as of species growing on highly disturbed areas increased in abundance in this layer.

The abundance of almost all woody species in the pollen diagram decreased, however the proportion of Pinus in the percentage pollen diagram increased (probably due to a sand admixture in the upper part of this layer). Among both charcoal and wood fragments, Pinus sylvestris dominated.

Layers 145 and 144

In both of these uppermost layers of the profile A20, the concentration of plant macro-remains decreased continually along with the number of determined taxa. The total number of pollen types also decreased. The only remaining aquatic species represented by macro-remains was Lemna, while the only remaining hygrophilous species was Ranunculus sceleratus. The composition of species typical for ruderal habitats was extended by Atriplex and cf. Galeopsis tetrahit, in addition to the Chenopodium species and Sambucus ebulus found also in previous layers. Among the weeds, the only remaining species were Fumaria officinalis and Polycnemum arvense (both in gradually decreasing concentrations). Glaucium corniculatum appeared for the first time in layer 144; two caryopses of Secale cereale were also found in this layer. The dominant pollen type was Polygonum aviculare, followed by species of the Poaceae, Chenopodiaceae, genus Artemisia and Rumex. The pollen of crops, ruderals, weeds and species of highly-disturbed sites increased in abundance. By contrast, the abundance of pollen of woody species was very low here.

Discussion

147 The permanently moist sediments comprise optimal conditions for the preservation of environmental information. However, the majority of wet sediments obtained until now in the city of Prague have originated chiefly from pits or wells (Opravil 1986, 1994, Čulíková 1987, 1998 a, b, 2001 a, b, 2005, 2008, 2010), which represent an inseparable mixture of material of various origins (both natural and anthropogenic). Indeed, it is difficult to find a suitable material in urban archaeology as the localities under focus are mostly completely built up. Any research of a former water reservoir located under recent buildings is rare (e.g. Hellwing 1997, Sasaki and Takahara 2011, Starec et al. 2012). Within the area of Prague, the only investigated wet sediments of more or less natural origin were those of the so-called Old Town defence system moat (Beneš et al. 2002) and the alluvial sediments of the Vltava River backwaters (Kozáková and Pokorný 2007, Čulíková 2010).

Time extent of the sedimentation record

The sediments investigated by this study were dated using the radiocarbon method (see Table 3). As for the lowermost part (first 6 cm) of the investigated profile A20, there were not enough macro-remains for dating. Moreover, this material contained almost no pollen, therefore we cannot make any conclusion concerning the age of the oldest part of the profile. The first date (from the sample located 10 cm above the bottom) corresponds approximately to the oldest known reference to the Rybník village (10th to 11th century). The second date (13th to 14th century) comes from the sample located 40 cm above the bottom. This material has probably sedimented shortly before the ending of the pond existence, which could be causally related to the demise of the Rybník village after the New Town foundation in 1348. This period was characterised by an extensive building activity in the neighbourhood.

The third date (10th to 11th century) was obtained from the clayey layer 161, set between the sandy layers above the aquatic sediments. According to the correspondence analysis, this sample is much more similar to the samples from layer 178 than to the samples originated from the adjacent layers. Therefore, we can assume that this material was the result of redeposition of older sediment, perhaps connected with the later increased building activity at the site. Another explanation takes into consideration an outwash from some higher positions of the drainage area. Either way, this peculiar date must lead us to extreme caution when interpreting the upper half of the profile.

Still, we consider the two dates from the lower half of the profile as being plausible. They are in accordance with our expectations, the succession between them is uninterrupted, and the fragments of ceramic found between them correspond to the 12th century. Besides, the emergence of Centaurea cyanus pollen by the end of the 10th century and the gradual growth of its curve in the course of the following centuries accords with the results of other investigations in the Czech Republic (Jankovská 1997, Kozáková et al. 2009).

Vegetation types

The classification of vascular plant species was based on the recent vegetation of the Czech Republic characterised by the diagnostic species (Chytrý and Tichý 2003). Following vegetational units were identified (see Table 4): aquatic vegetation, hygrophilous herbaceous

148 vegetation, hygrophilous woodland vegetation, grassland vegetation (meadows, pastures and dry grasslands), annual vegetation of ruderalised sites, and weeds.

Aquatic vegetation

Among aquatic macrophytes, the following species were documented: Ceratophyllum demersum, Lemna minor, Myriophyllum spicatum, Potamogeton crispus, Potamogeton cf. natans, and Zannichellia palustris. Classes of Lemnetea and Potametea (Nymphaeion albae, Magnopotamion, Parvopotamion) could be identified according to the diagnostic species. These are common vegetation types of standing water extending nearly all over the world. Nevertheless, the macro-remains of these species are not very common in other documented archaeobotanical assemblages from Prague. For a more detailed description of the development of the water environment of this site see the following section.

Hygrophilous herbaceous vegetation

The following species of hygrophilous herbs were found: Alisma plantago-aquatica, Bidens cernua, Bidens cf. tripartita, Butomus umbellatus, Eleocharis palustris, Gnaphalium uliginosum, Lycopus europaeus, Lythrum salicaria, Persicaria hydropiper, Ranunculus sceleratus, Rorippa palustris, Saggitaria sagittifolia, Sparganium erectum, Sparganium emersum, and Typha latifolia. Classes of Isoeto-Nanojuncetea (Eleocharition ovateae), Phragmito-Magnocaricetea (Phragmition communis, Oenanthion aquaticae) and Bidentetea tripartitae (Bidention tripartitae) could be identified according to the diagnostic species.

The species of hygrophilous vegetation culminated (both quantitatively and qualitatively) in the lower half of the profile, and showed a decreasing tendency. This trend was prominent both in the macro-remains and pollen. These plants were probably growing on sites connected immediately with the pond. Similar vegetation was detected near a backwater in Malá Strana (Čulíková 2010, Kozáková and Pokorný 2007). Hovewer, Eleocharis and Lycopus have been found in nearly all sites in Medieval Prague, including pits (Opravil 1986, 1994, Čulíková 1998a, b, 2001a, b, 2005, 2010).

Hygrophilous woodland vegetation

The following taxa of hygrophilous woodland vegetation were found: Salix, Alnus cf. glutinosa, Aegopodium podagraria, Chaerophyllum hirsutum, Fraxinus exelsior, cf. Myosoton aquaticum (Cerastium fontanum-group), Rubus idaeus, Rumex crispus/obtusifolius, Sambucus nigra, and Urtica dioica. Classes of Salicetea purpureae (Salicion triandreae) and Querco-Fagetea (Alnion incanae) could be identified according to the diagnostic species.

These species were represented mostly by pollen or wood (charcoal). Only Rumex, Urtica, Rubus idaeus and Sambucus nigra (edible fruits) were also found in the macro-remains. This could be intepreted as a consequence of a situation when this type of vegetation was growing nearby, but not directly on the site.

Table 4 Vegetation units identified according to (recent) diagnostic species (classification of plant taxa into syntaxa follows Chytrý & Tichý 2003) based on finds of macro-remains.

149 Ecological groups 1 2 3 4 5 6 7 8 9 Syntaxa A B C D E F G H I J K L M N O P Q R Aethusa cynapium ...... o . . . . Agrostemma githago ...... x . . . . Ajuga cf. reptans ...... o + Alisma plantago-aquatica . . + . o ...... Anagallis arvensis ...... + + . . . . Anthemis arvensis ...... o . . . . Anthemis cotula ...... x . . . . . Aphanes arvensis ...... o . . . . Arenaria serpyllifolia agg...... o . . . + . . . . Atriplex sp...... o . . . . . Bidens cernua . . . . . o ...... Bidens cf. tripartita . . + . . o ...... Bupleurum rotundifolium ...... x . . . . Capsella bursa-pastoris ...... + + . . . . Carex cf. hirta ...... o ...... Carex cf. pallescens ...... + ...... Carex leporina ...... o ...... Centaurea cyanus ...... + . . . . Ceratophyllum demersum o o ...... cf. Galeopsis tetrahit ...... + . . . . cf. Myosoton aquaticum ...... + . . . cf. Sparganium erectum . . . o ...... cf. Thalictrum flavum ...... + ...... cf. Typha latifolia . . . o ...... Chenopodium album agg...... + + . . . . Chenopodium ficifolium ...... o . . . . . Chenopodium hybridum ...... + . . . . . Chenopodium polyspermum ...... o . . . . . Cirsium cf. palustre ...... o ...... + . . Corylus avellana ...... + Dianthus armeria/deltoides ...... o ...... Echinochloa crus-galli ...... o . . . . . Eleocharis palustris agg. . . . . o ...... Euphorbia helioscopia ...... + + . . . . Fallopia convolvulus ...... + + . . . . Filipendula ulmaria ...... o ...... + + . . Fumaria officinalis ...... o . . . . Geranium molle/collumbinum ...... + ...... Glaucium corniculatum ...... x . . . . Herniaria glabra ...... o ...... Hypericum perforatum ...... + o ...... Juncus sp. . . + . . . o ...... Lamium cf. amplexicaule ...... o + . . . . Lapsana communis ...... o + . . . . Lemna minor/gibba o + . o ...... Leontodon autumnalis ...... + ...... Ecological groups 1 2 3 4 5 6 7 8 9 Syntaxa A B C D E F G H I J K L M N O P Q R Leucanthemum vulgare agg...... o ...... Lychnis flos-cuculi . . . . . o ...... Lycopus europaeus . . . o o + ...... + . . Myriophyllum cf. verticillatum . + ......

150 Myriophyllum spicatum . + ...... Neslia paniculata ...... o . . . . Papaver cf. rhoeas ...... + . . . . Persicaria hydropiper . . . . . o ...... Persicaria lapathifolia agg. . . + . . + ...... + + . . . . Persicaria maculosa ...... o . . . . Picea abies ...... + Picris cf. hieracioides ...... + ...... Polycnemum arvense ...... x . . . . Polygonum aviculare agg...... o + + . . . . Potamogeton cf. natans . + ...... Potamogeton crispus . + ...... Potentilla anserina ...... o ...... Prunella vulgaris ...... o ...... Ranunculus acris ...... o + . . + ...... Ranunculus aquatilis . o ...... Ranunculus cf. fluitans . + ...... Ranunculus repens . . . . . + o . . . . + + . + + . . Ranunculus sceleratus . . o . . o ...... Rorippa palustris . . o . . o ...... Rubus idaeus ...... o + Rumex cf. crispus ...... + . . . . Rumex cf. obtusifolius ...... + . . . Sambucus ebulus ...... x . . . . . Sambucus nigra ...... + . o + Scirpus sylvaticus ...... o ...... + . . Scleranthus anuus ...... + . . . + . . . . Setaria cf. viridis ...... + . . . . . Setaria pumila ...... o . . . . . Lychnis cf. viscaria ...... + ...... o Solanum dulcamara ...... o . . Stachys cf. annua ...... o . . . . Stellaria graminea ...... o ...... Stellaria media agg...... o + . . . . Taraxacum officinale agg...... + . . . . + + + . . . . Thlaspi arvense ...... + + . . . . Urtica dioica . . . + + ...... + + o + Urtica urens ...... o . . . . . Vaccaria hispanica ...... x . . . . Valerianella dentata ...... o . . . . Zannichellia palustris . + ......

Ecological groups and the syntaxa relating to the various groups 1 Aquatic vegetation. A, Lemnetea; B, Potametea 2 Hygrophilous herbaceous vegetation. C, Eleocharion ovatae; D, Phragmition communis; E, Oenanthion aquaticae 3 Ruderalised hygrophilous habitats. F, Bidention tripartitae 4 Grassland vegetation (meadows, pastures and dry grasslands). G, Calthion; H, Arrhenatherion; I, Plantagini- Festucion ovinae; J, Festuco-Brometea; K, Violion caninae 5 Trampled habitats. L, Plantaginetea majoris 6 Annual vegetation of ruderalised sites. M, Chenopodietea (Fumario-Euphorbion, Spergulo-Oxalidion, Panico- Setarion) 7 Crop weeds. N, Secalietea (Caucalidion lappulae, Sherardion, Veronico politae-Taraxacion, Aphanion) 8 Hygrophilous woodland vegetation. O, Salicion triandrae; P, Alnion glutinosae; Q, Alnion incanae 9 Trees. R, Querco-Fagetea + - Taxon occurs in the group

151 o - Taxon has its main occurrence in the group x - Taxon is considered to have occurred in the group in the pastDry grasslands

Wet meadows

The following species (or pollen types) indicating wet meadows were found: Angelica sylvestris, Alchemilla pentaphyllea, Alnus, Caltha palustris, Carex cf. hirta, Carex cf. pallescens, Chaerophyllum hirsutum, Cichorium intybus type, Cirsium palustre type, Filipendula ulmaria, Geranium molle type, Heracleum sphondylium, Lathyrus, Leucanthemum vulgare aggr., Lychnis flos-cuculi, Lysimachia vulgaris type, Ranunculus acris, Ranunculus repens, Rubiaceae, Rumex acetosa type and Scirpus sylvaticus. The class Molinio- Arrhenatheretea (Calthion) could be identified according to the diagnostic species. Also Stellaria graminaea, the species with a wide ecological valence from dry to wet meadows was among the list of grassland species.

Both the macro-remains and pollen of this group showed a decreasing tendency in the profile. Many of them were also found in the backwater in Mala Strana (Čulíková 2010, Kozáková and Pokorný 2007), as well as in the drainage ditch of the Old Town defence system (Beneš et al. 2002), and in the Prague Castle (Kozáková and Boháčová 2008). Moreover, the following taxa were also found in other archaeological sites in Medieval Prague: Caltha palustris, Lychnis flos-cuculi, Ranunculus, Scirpus sylvaticus, Stellaria graminaea and Cirsium. It is thus possible to conclude that meadows were probably quite widespread in Prague suburbs (mostly around the river and its tributaries) and the material of the meadows (hay) was also manipulated in households (or it could have been transported by animals and deposited in cesspits in the form of dung).

The occurrence of some rare pollen types (e.g Lychnis viscaria type, Anthericum type, Saxifraga oppositifolia type, Thesium type, Aster tripolium type, Helianthemum, Lithospermum arvense, Medicago lupulina type, Echium type, Teucrium type and Verbascum type) in the pollen spectrum would imply that xerophilous grasslands probably developed in the close vicinity of the investigated water reservoir. The area of Prague has many places where biotopes mainly belonging to the class Festuco-Brometea could occur. These habitats were probably used as pastures in the past. The presence of grazing indicators (e.g. pollen of Poaceae, Plantago lanceolata, Calluna vulgaris, Jasione montana type, and charcoal from Juniperus) in the pollen spectrum would support such a scenario (Behre 1981). Those habitats with remaining dry grassland vegetation within Prague are nowadays generally under the status of nature reserves (Dostálek and Frantík 2008). We could suppose that such habitats were more common within the area before the New Town of Prague was established.

Annual vegetation of ruderalised sites

The following species of ruderal vegetation were found: cf. Capsella bursa-pastoris, Chenopodium album aggr., Chenopodium hybridum, Chenopodium polyspermum, Echinochloa crus-galli, Euphorbia helioscopia, Lapsana communis, Polygonum aviculare aggr., Setaria pumila, Stellaria media aggr., and Thlaspi arvense. The present day class Chenopodietea

152 (Fumario-Euphorbion and Panico-Setarion) could be identified according to the diagnostic species.

This group of vegetation shows an increasing tendency, culminating in the upper half of the profile. It corresponds with the pollen curve of trampled vegetation indicators, and this trend is at the same time opposite to the trend of semi-natural types of vegetation mentioned above. When comparing with other sites, the macro-remains of these species are nearly ubiquitous both in archaeological sites and in natural sediments, so this vegetation probably grew inside the town and around the villages, as well as along watercourses. In contrast, perennial ruderals were very scarce on this site. Nevertheless, their particular occurrence in the Old Prague trade centre Ungelt (Opravil 1986) indicates the different environment in the urban and trade centre of that time.

Weeds

The following weeds were found: Anagallis arvensis, Aethusa cynapium, Agrostemma githago, Anthemis arvensis, Aphanes arvensis, Arenaria serpyllifolia aggr., Bupleurum rotundifolium, cf. Capsella bursa-pastoris (Brassicaceae), Centaurea cyanus, Chenopodium album aggr., Euphorbia helioscopia, Fallopia convolvulus, Fumaria officinalis, Glaucium corniculatum, Lapsana communis, Lithospermum arvense, Microrrhinum minus, Neslia paniculata, Papaver cf. rhoeas, Plantago lanceolata, Plantago major, Polycnemum arvense, Polygonum aviculare aggr., Rumex crispus/obtusifolius, Scleranthus anuus, Stachys cf. annua, Stellaria media aggr., Thlaspi arvense, and Valerianella dentata. The present day class Secalietea (Caucalidion lappulae, Sherardion, Aphanion, and Veronico politae-Taraxacion) could be identified according to the diagnostic species. However, some of these weedy species are no longer present in present-day fields, although they were very frequent in Medieval fields (i.e. Agrostemma githago, Glaucium corniculatum, Bupleurum rotundifolium, Polycnemum arvense).

Both the number of species and quantity of weeds increased in time (both pollen and macro-remains), along with the pollen of cereals. Many of these weed species are ubiquitous in archaeological sites (Aethusa cynapium, Anagallis arvensis, Anthemis arvensis, Arenaria serpyllifolia, Centaurea cyanus, Fumaria officinalis, Lithospermum arvense, Neslia paniculata, Fallopia convolvulus, Scleranthus annuus, and Valerianella dentata). The possible explanation is that they frequently entered settlements along with the field products brought in. Nevertheless, some weedy species often found in cesspits were not encountered on this site (e.g. Caucalis platycarpos, Galium spurium, Lithospermum arvense, Adonis aestivalis, Sinapis arvensis). On the other hand, some weedy species were almost only found here (e.g. Anthemis cotula, Vaccaria hispanica, and Scleranthus annuus) though some of them were also encountered at Mala Strana.

Environmental changes

Terrestrial vegetation

153 During the period between the end of 10th to approximately the middle of 14th century, rather extensive changes of plant species composition took place in the region of Prague. The changes observed at the investigated site clearly correspond with the general trends in landscape management reconstructed by Kozáková et al. (2009). According to this reconstruction, a fine mosaic of habitats existed in Prague before the end of 12th century. Subsequently, the environmental diversity decreased considerably during the following centuries. This trend was explained as a result of increasing ruderalization and intensification of the land use.

In the site investigated, the field indicators (both cereals and field weeds) increased over time (Fig. 4), whereas the proportion of broad-leaf trees and shrubs decreased (Fig. 5). This trend in the pollen spectra could be interpreted as a manifestation of the gradual enlargement of land under the plough. At the same time, the proportion of ruderal plants increased continually both in the pollen and macro-remains spectra. The trend in grassland indicators is much less clear; however, this type of vegetation seems to decrease slightly with time. A gradual decline in the semi-natural hygrophilous vegetation was accompanied by an opposite tendency in trampled vegetation (Polygonum aviculare, Rumex acetosella). This trend, along with the distinct increase of landscape ruderalisation, seems to indicate the gradual intensification of human activity around the water pool. A similar intensification of anthropogenic influence is clearly visible in the development of the aquatic environment of the pool, which has been documented thoroughly (see below).

154

Fig 4 Trends in development of vegetation units (as recorded in the A20 profile) expressed by both sums of macro- remains (histograms) and percentage of pollen within the total pollen spectrum (curves). Dark grey curves represent total percentage of pollen types assigned to individual vegetation units; whereas the light grey area within the curve represent the proportion of selected pollen type within the unit. Total number of determined pollen types is expressed by the Shannon index (dimensionless quantity). The depth from the surface (y axis) is expressed in cm. Calibrated radiocarbon dates (cal yr AD) are plotted along the y axis

155

Fig 5 Occurrence of woody plants in the A20 profile. Mixed diagram of both pollen (percentage within the total pollen spectrum expressed by curves, grey areas depict the same values exaggerated ten-fold) and charcoal/wood (count of pieces expressed by histograms). The depth from the surface (y axis) is expressed in cm. Calibrated radiocarbon dates (cal yr AD) are plotted along the y axis

156 The area of Prague belongs to the Bohemian Thermophytic region, the Prague Basin subunit of the phytogeographical region called The Prague Plateau (Skalický 1997). The potential vegetation of this site is Hercynian oak-hornbeam forest (Carpinion), acidophilous oak forest (Genisto germanicae-Quercion) and ash-alder alluvial forest (Alnenion glutinoso- incanae) (Neuhäuslová et al. 1998, Moravec and Neuhäusl 1991). According to the results of pollen, wood and charcoal analyses, all of these types of forest vegetation were probably present in the vicinity of the study site. An analogous species composition has been recorded from other Prague Medieval sites (Beneš et al. 2002, Kozáková and Pokorný 2007, Kozáková et al. 2009, and Novák et al. 2012). The results originating from both charcoal and wood analyses revealed a lower number of species than the pollen analysis. Some rare shrub species (e.g. Viburnum opulus type, Cornus mas type, Juglans regia type) were recorded only through the pollen analysis. Such differences between the pollen and the charcoal/wood analyses reflect the various source areas. The species composition of charcoal fragments reflects the use of firewood originating in close proximity of the site (Théry-Parisot et al. 2010, Novák et al. 2012). In contrast, the species composition of wood often reflects local vegetation cover, e.g. willow branches, or selectively collected construction material (fir, spruce).

Aquatic environment

Using a series of various methods, it was possible to make a detailed reconstruction of the aquatic environment of the site and its gradual changes. The changes of water quality were documented by a series of organisms: water macrophytes (Ceratophyllum demersum, Lemna minor/gibbaI, Myriophyllum spicatum, Potamogeton, Sparganium, and Zannichellia pallustris), green algae (Pediastrum simplex, Tetraedron minimum, Scenedesmus), cladocera (ephipia of D. magna and D. cf. pulex), fish (Cyprinidae: Tinca tinca, Leuciscus cephalus, Alburnus alburnus, Rutilus rutilus, and Perciformes), rotifers and diatoms of various species. On the grounds of the diatom composition, an index of saprobity was calculated, which could be used as a measure of water quality (Fig. 6).

According to the index of saprobity (1.2), the base of the profile (layers 181 and 182) was the result of sedimentation in considerably oligosaprobic conditions. The index of saprobity then gradually increased, exceeding 2 within layer 179, followed by the absolute disappearance of diatoms in layer 178. The rapid and almost synchronous decrease of green algae, diatoms and fishes in layer 178 implies that it was probably the response of aquatic organisms to some changes in the environment. The water was obviously greatly shaded by a thick layer of both Lemna gibba/minor and Potamogeton natans. This could explain the decrease of green algae but not also the decrease of diatoms occurring in early spring and late autumn, when summer shadowing by floating plants should be irrelevant. Alternatively, the high content of dissolved nutrients, which was probably connected with anoxia, could have been a major culprit causing the disappearance of both diatoms and fishes. Pollution of the water by organic waste (e.g. excrement) could be the principal reason for the rapid decrease in water quality. This interpretation is supported by appearance of the intestinal parasite Trichuris (not the human one) eggs shortly before the diatoms disappeared. Other waste indicators found in these two layers were as follows: seed of cucumber, seed of grape, pig bone and several pieces of ceramics.

157

Fig 6 Quantification of proxy-data indicating the aquatic environment (as recorded in the A20 profile). The depth from the surface (y axis) is expressed in cm. Calibrated radiocarbon dates (cal yr AD) are plotted along the y axis. The count of macro-remains (macrophyta, algae, resting eggs of rotifers, Trichuris eggs as well as fish remains) is expressed in pieces (x axis). Aquatic macrophyta are represtnted by both sum of macro-remains (M) and percentage of pollen (P). Estimated abundance of mussels and Daphnia are plotted as black dots; the size of dots (small, medium, big) corresponds to the estimated quantification. The level of aquatic pollution is expressed by index of saprobity (dimensionless quantity) calculated according to diatom spectrum

According to the environmental demands of the fish (Table 2), the aquatic environment of the pond was probably rather diversified - the determined fish species occur both in lotic (i.e. flowing) and stagnant water (e.g. ponds and pools with muddy substrates). Tench and chub tolerate a lower amount of oxygen dissolved in water and prefer sites overgrown by vegetation. Species of the order Perciformes could occur in various types of water pools (e.g. lagoons, lakes of all types, or slow to medium flowing watercourses).

As opposed to the lower half of the profile, the intrerpretation of the upper part is much less clear. We assume that the layers 180, 179 and 178 definitely represent sedimentation in water reservoir. On the other hand, the overlying sediments (layers 162, 161, 145 and 144) are characterised by a rapid decrease in the number of remains of aquatic organisms. This could be due to a deterioration of preservation conditions. Besides, the psammitic (fine sandstone)

158 character of the sediment of these layers implies an increased influx of material, which could be connected to an increased risk of contamination. Moreover, the possibility of the redeposition of an older material (the 14C date within the layer 161, see above) should be taken into consideration.

There are two possible explanations of this phenomenon. One of the possibilities is that the material of the layer 161 is the result of redeposition of older sediment, perhaps connected to subsequent increase of building activity at the site. In that case all the material from the upper part of the profile would be heavily contaminated. Alternatively, the older material encountered in the layer 161 could have been washed away from another location and later settled on the investigated site, similarly to the sandy sediments of the layers 162, 145 and 144, which are probably the result of an outwash from some higher positions of the drainage area. In that case the assamblages of bio-facts encountered in the upper part of the profile would represent a mixture of material of both local and regional origin. Nevertheless, we suppose that the second scenario is more probable because of continuous succession of anthropogenic indicators in both macro-remains and pollen spectra. On the other hand, it is not possible to determine whether the water reservoir merely changed its character after the middle of 14th century or it disspeared and only a wet place with seasonal water influx remained.

Origin and function of the pond

Origin

The existence of the water reservoir is obviously related to the strong springs rising on the boundary between the permeable gravel of the Vinohrady riverine terrace and the underlying impermeable clay and bedrock (Zavřel 1999, 2009 - unpublished reports). The groundwater emerging from the base of this terrace also probably flowed inside the permeable layers of the secondary transported sands, afterwards seeping out onto the surface. In addition to that, the water from the terrace could also have flowed above the surface creating occasional small watercourses. For all these reasons the site and its near surroundings had been characterised for a long time by a high level of groundwater (Zavřel 2009, unpublished report) and numerous wetlands and springheads (for more details to the hydrology of the New Town of Prague, see Zavřel (2006)).

The most probable explanation of the pond origin is that this naturally wet place was dammed artificially. However, this could not be proved definitively, as the supposed dam position lies outside the excavation area. Nevertheless, a 50 m section of a brook was found in the close vicinity of the investigated area (Fig. 2) (Kašpar 2003). There could be two possible interpretations of the mutual relationship between these two features (the pond and the brook): According to the slope direction, either the brook flowed out of the pond, or the brook had no connection with the pond (in which case the pond would probably be very small).

Function

159 Knowledge of carp breeding and fishpond dam construction came to the Czech lands from West Europe (Makowiecki 2001), probably with monks from Germany. Before this time, so called “vivaria piscium”, i.e. artificial water pools (stews, “stavy”) used for storage of captured fishes (Čítek et al. 1998), were constructed here, probably already since the 8th to 9th centuries (Andreska 1987). Their existence in Moravia near Olomouc has been supported by a document from 1078 (Hurt 1960). The oldest reference concerning a fishpond in Bohemia is a mention in the Chronica Bohemorum, in the part relating to the foundation of the Sázava monastery (Kosmas 2011) in the 11th century. In 1227, the king Ottokar II of Bohemia authorized the construction of fishponds. In the 14th century, fish farming was already an important economic activity in the Czech lands. Since the first half of the 14th century, small ponds were built in the middle of each village, serving both for fish farming, and as fire water reservoirs (Čítek et al. 1998).

According to the fish species composition and the dating of the sediments, we could assume that the pond could have served for the retention of water and/or storing of fish, but not for fish breeding. During the earlier phases of the pond’s existence, the water was clean enough for drinking; however, it was later heavily contaminated by organic material. Still, it could have served as a water reservoir for various handicrafts, as, for example, brickmaking, pottery, ironworking or leather processing.

The further history of the pond remains unclear. The uppermost layers of the sedimentary sequence had been destroyed during the construction of the basement. However, the existence of three ponds belonging to a farmstead nearby was documented even in 1428 (Tomek 1892). According to historical maps from 1791 (Herget’s plan) and from 1816 (Jüttner’s plan), this part of the New Town of Prague remained undeveloped (at least not built on) and the ground concerned was a part of some open areas or garden plots.

Conclusions

The results of this study allow the following conclusions:

1. According to the results of the environmental analyses, the material of the lower part of the profile is definitely a result of sedimentation on the bottom of a water reservoir. The duration of the water body was estimated for the period from the end of 10th century to approximately half of 14th century according to radiocarbon dating. Several species of freshwater fishes were determined according to bones/scales, but no remains of carp were found. Therefore we can conclude that this was probably a kind of water reservoir containing fishes; however it was not a typical fishpond intended for carp farming. 2. The origin of the water reservoir was connected with rich water springs having their source on the slope of the riverine terrace. The pond itself was probably constructed artificially, or alternatively a shallow natural pool could have been deepened and extended. Nevertheless, because the extent of the investigation area was too small, any remains of the dam structure were not documented. 3. The water quality was estimated according to proxy data. We assume that the water was originally relatively clean and the base of the profile was the result of sedimentation in considerably oligosaprobic conditions. However, increasing degree of organic pollution was documented in subsequent layers, caused probably by human influence (occasional waste disposal, defecation of domestic animals).

160 4. The character of the surrounding vegetation was reconstructed according to both pollen and macro remains data. Originally, the water body was surrounded by vegetation of semi- natural character (i.e. hygrophilous herbaceous vegetation directly on the banks as well as wet meadows, pastures and willow shrubs in the proximity). Later, gradual changes in the surrounding environment were documented. Increasing ruderalisation and intensification of human impact in closer proximity were manifested by indicators of trampled vegetation and nitrophilous plant species. Additionally, increasing proportion of cereals and decreasing proportion of trees in the pollen diagram indicated significant changes taking place on the landscape level. Acknowledgement This research was supported by the project GACR 13-11193S and "PAPAVER - Centre for Human and Plant Studies in Europe and Northern Africa in the Postglacial Period", reg. No. CZ.1.07/2.3.00/20.0289 and co-financed by the European Social Fund and the state budget of the Czech Republic. We are grateful to Dr. Helena Svobodová- Svitavská for help with ambiguous pollen determinations and to Dr. Miloslav Devetter for valuable comments on rotifers eggs determination. We wish to thank Anna Svatušková for help during the excavation and Steve Ridgill for making improvements to our English.

References

Andreska J (1987) Rybářství a jeho tradice [Fishing and its tradition]. Státní zemědělské nakladatelství, Praha

Baruš V, Oliva O (1995) Mihulovci (Petromyzontes) a ryby (Osteichthyes) 2 [Lampreys (Petromyzontes) and fishes (Osteichthyes) 2]. Academia, Praha

Battarbee R (1986) Diatom analysis. In Berglund B.E. (ed.), Handbook of Holocene Paleoecology and Paleohydrology. Wiley and Sons, Chichester

Behre KE (1981) The interpretation of antropogenic indicators in pollen diagrams. Pollen et Spores 23:225-245

Beran L (1998) Vodní měkkýši ČR [Aquatic molluscs of the Czech Republic]. ČSOP, Vlašim

Beug HJ (2004) Leitfaden der Pollenbestimmung für Mitteleuropa und angrenzende Gebiete. Verlag Friedrich Pfeil, München

Birks HH, Birks HJB (2006) Multiproxy studies in palaeolimnology. Veget Hist Archaeobot 15:235-251

Bosi G, Bandini Mazzanti M, Florenzano A, Massamba N'siala I, Pederzoli A, Rinaldi R, Torri P, Mercuri AM (2011) Seeds/fruits, pollen and parasite remains as evidence of site function: Piazza Garibaldi - Parma (N Italy) in Roman and Mediaeval times. Journal of Archaeological Science 38:1621–1633

161 Cappers RTJ, Bekker RM, Jans JEA (2006) Digitale Zadenatlas van Nederland [Digital Seed Atlas of the Netherlands]. Barkhuis & Groningen University Library, Groningen

Chytrý M, Tichý L (2003) Diagnostic, constant and dominant species of vegetation classes and alliances of the Czech Republic: a statistical revision. Masarykova Universita, Brno

Čítek J, Krupauer V, Kubů F (1998) Rybnikářství [Fish farming]. Informatorium, Praha

Čulíková V (1987) Zajímavý nález rostlinných makrozbytků ze středověké Prahy [An interesting find of plant macro-remains from medieval Prague]. Archeologické rozhledy 39:445–452

Čulíková V (1998a) Rostlinné makrozbytky z raně středověkých sedimentů na III. nádvoří Pražského hradu [Plant macro-remains from early medieval sediments in the 3rd courtyard of Prague castle]. Archaeologica Pragensia 14:329–341

Čulíková V (1998b) Výsledky analýzy rostlinných makrozbytků z lokality Praha 1 – Malá Strana, Tržiště čp. 259/III (Hartigovský palác) [Results of the analysis of plant macro-remains from the locality Prague 1-Lesser Town, Tržiště no. 259/III (Hartigovský palác palace)]. Archeologica Pragensia 14:291–316

Čulíková V (2001a) Rostlinné makrozbytky z lokality Praha 1 – Malá Strana, Malostranské nám. čp. 258/III (Lichtenštějnský palác) [Plant macro-remains from the locality Prague 1- Lesser Town, Malostranské nám. no. 258/III (Lichtenštejnský palác palace)]. Medievalia archaeologica 3:137–166

Čulíková V (2001b) Rostlinné makrozbytky z pěti středověkých lokalit při obvodu centrální části Pražského hradu [Plant macro-remains from five medieval localities around the perimeter of the central part of Prague castle]. Medievalia archaeologica 3:303–327

Čulíková V (2005) Rostlinné makrozbytky z raně středověké lokality Mostecká – Josefská ul. (dřevěná cesta), Praha 1 – Malá Strana [Plant macro-remains from the early medieval site at Mostecká – Josefská str. (wooden causeway), Prague 1-Lesser Town]. Archeologica Pragensia 17:137–169

Čulíková V (2008) Ovoce, koření a léčiva z raně novověké jímky hradčanského špitálu [Fruits, spices and medicaments from the post-medieval cesspit of the hospital at Prague-Hradčany]. Archeologické rozhledy 60:229–260

Čulíková V (2010) Středověká údolní niva Vltavy v Praze na Malé Straně (Valdštejnská čp. 154/III, Kolovratský palác) [Medieval floodplain of the Vltava river in the neighbourhood of the Lesser Town in Prague (Valdštejnská st. no. 154/111, Kolovrat palace)]. Archeologické rozhledy 62:72–116

Dostálek J, Frantík T (2008) Dry grassland plant diversity conservation using low-intensity sheep and goat grazing management: case study in Prague (Czech Republic). Biodiversity: Research and Conservation 17:1439–1454

162 Fægri K, Iversen J (1964) Textbook of pollen analysis, 2nd edition. Scandinavian University Books, Copenhagen

Fott B (1954) Pleurax, synthetická pryskyřice pro preparaci rozsivek [Pleurax, the synthetic resin for diatoms preparation]. Preslia 26:163–194

Gaillard MJ (2007) Pollen Methods and Studies - Archaeological Applications. In Elias S.A. (ed) Encyclopedia of Quaternary Science. 2570-2595, Elsevier, Oxford

Grimm EC (1987) CONISS: A FORTRAN 77 program for stratigraphically constrained cluster analysis by the method of incremental sum of squares. Computers & Geosciences 13:13–35

Grimm EC (2011) Tilia. Version 1.5.12. Illinois State Museum, Springfield, Illinois, USA

Heiss AG (2000) Anatomy of European and North American woods - an interactive identification key. URL: http://www.holzanatomie.at/

Hellwig M (1997) Plant remains from two cesspits (15th and 16th century) and a pond (13th century) from Götingen, southern Lower Saxony, Germany. Veget Hist Archaeobot 6:105–116

Hrdlička L (1984) Nástin vývoje reliéfu historického jádra Prahy ve středověku [A sketch of the development of medieval historical centre of Prague]. Archaeologica Pragensia 5:197–209

Hrdlička L (1997) Od spontánního vývoje k regulaci terénního reliéfu Starého Města pražského [From spontaneous terrain development to regulation of relief in the Old Town of Prague]. In Kubková J., J. Klápště, M. Ježek & P. Meduna (eds) Život v archeologii středověku. 246-252, Archeologický ústav AV ČR, Praha, v.v.i., Praha

Hurt R (1960) Dějiny rybnikářství na Moravě a ve Slezsku [History of fish farming in Moravia and Silesia]. Krajské nakladatelství v Opavě, Opava

Jankovská V (1997) Výsledky pylových analýz z lokality: Praha 1 – Malá Strana, Tržiště 259/III [Results of pollen analysis from Lesser Town, Tržiště Street 259/III, Prague]. In Kubková J., J. Klápště, M. Ježek & P. Meduna (eds) Život v archeologii středověku. 299–308, Archeologický ústav AV ČR, Praha, v.v.i., Praha

Jankovská V, Pokorný J (2002) Palaeoecology of a medieval fishpond system (Vajgar, Czech Republic). Folia Geobotanica 37:253-273

Juggins S (2009) Rioja: analysis of Quaternarry science data, R package version 0.5-6. http://cran.r.-project.org/package=rioja

Kašpar V (2003) Rotunda sv. Longina a osada Rybník na Novém Městě pražském [St. Longin's Rotunda and the Rybnik village in the New Town of Prague]. Gorgoneion 1:15–42

Kašpar V (2007) Politických vězňů čp. 913/II. In Dragoun Z. (ed) Pražský sborník historický 35. 352–353, Praha

163 Kaštovský J, Kočár P, Kočárová R, Pokorný P, Beneš J, Starec P (1999) Předběžné poznatky o některých vodotečích na území Starého a Nového Města pražského [Preliminary knowledge of some watercourses in the area of the New and Old Towns of Prague]. Archaeologia Historica 24:143–151

Katz NJ, Katz SV, Kipiani MG (1965) Atlas and keys of fruits and seeds occurring in the quaternary deposits of the USSR. Nauka, Academy of Science of the USSR, Moscow

Komárek J, Jankovská V (2001) Review of the Green Algal Genus Pediastrum; Implication for Pollen-Analytical Research. Gebrüder Borntraeger Verlagsbuchhandlung, Berlin, Stuttgart

Kosmas (2011) Kronika Čechů [Chronicle of the Bohemians]. Hrdina K, Bláhová M (translation). Argo, Praha

Kozáková R, Boháčová I (2008) Přírodní prostředí Pražského hradu a jeho zázemí v raném středověku – výpověď pylové analýzy sedimentů ze III. nádvoří [Environment of Prague Castle and its vicinity – notice of pollen analysis from Early Medieval sediments from the 3rd courtyard]. Archeologické rozhledy 60:547–564

Kozáková R, Pokorný P (2007) Dynamics of the biotopes at the edge of a medieval town: pollen analysis of Vltava river sediments in Prague, Czech Republic. Preslia 79:259–281

Krammer K, Lange-Bertalot H (1986) Bacillariophyceae (Naviculaceae). In Ettl H., J. Gerloff, H. Heinig & D. Mollenhauer (eds), Süsswasserflora von Mitteleuropa 1, Gustav Fischer Verlag, Stuttgart

Krammer K, Lange-Bertalot H (1988) Bacillariophyceae (Bacillariaceae, Epithemiaceae, Surirellaceae). In Ettl H., J. Gerloff, H. Heinig & D. Mollenhauer (eds), Süsswasserflora von Mitteleuropa 2, Gustav Fischer Verlag, Stuttgart

Krammer K, Lange-Bertalot H (1991a) Bacillariophyceae (Achnanteaceae, Kritische Ergänzungen zu Navicula (Lineolateae) und Gomphonema Gesamtliteraturverzeichnis). In Ettl H., J. Gerloff, H. Heinig & D. Mollenhauer (eds), Süsswasserflora von Mitteleuropa 4, Gustav Fischer Verlag, Stuttgart

Krammer K, Lange-Bertalot H (1991b) Bacillariophyceae (Centrales, Fragilariaceae, Eunotiaceae). In Ettl H., J. Gerloff, H. Heinig & D. Mollenhauer (eds), Süsswasserflora von Mitteleuropa 3, Gustav Fischer Verlag, Stuttgart

Kreuz A, Schäfer E (2002) A new archaeobotanical database program. Veget Hist Archaeobot 11:177–179

Kubát K, Hrouda L, Chrtek jun. J, Kaplan Z, Kirschner J, Štěpánek J (2002) Klíč ke květeně České republiky [Field guide to flora of the Czech Republic]. Academia, Praha

164 Legendre P, Legendre L (1998) Numerical Ecology. 2nd English edition. Elsevier, Amsterdam

Makowiecki D (2001) Some remarks on medieval fishing in Poland. In Buitenhuis H. & W. Prummel (eds) Animals and Man in the Past. 236–241, ARC-Publicatie 41, Groningen

Mencl V (1969) Praha. Odeon, Praha

Moravec J, Neuhäusel R, Blažková D, Husová M, Kolbek J, Krahulec F, Neuhäuslová-Novotná Z (1991) Přirozená vegetace území hlavního města Prahy a její rekonstrukční mapa [Natural vegetation of the capital city Prague and its reconstruction map]. Academia, Praha

Neuhäuslová Z (ed) (1998) Mapa potenciální přirozené vegetace České republiky [Map of potential natural vegetation of the Czech Republic]. Academia, Praha

Novák J, Lisá L, Pokorný P, Kuna M (2012) Charcoal analyses as an environmental tool for the study of Early Medieval sunken house infills in Roztoky near Prague, Czech Republic. Journal of Archaeological Science 39:808–817

Oksanen J, Blanchet FG, Kindt R, Legendre P, O'Hara RB, Simpson GL, Solymos P, Stevens MH, Wagner H (2010) Vegan: Community Ecology Package. R package version 1.17-2. http://cran.r.-project.org/package =vegan

Opravil E (1986) Rostlinné makrozbytky z historického jádra Prahy [Plant macro-remains from the historical centre of Prague]. Archaeologica Pragensia 7:237–271

Opravil E (1994) Příspěvek k poznání rostlinných makrozbytků ze staré Prahy [A contribution to the knowledge of plant macro-remains from historical Prague]. Archeologické rozhledy 46:105–114

Pfleger V (1988) Měkkýši [Molluscs]. Artia, Praha

Pokorná A, Dreslerová D, Křivánková D (2011) Archaeobotanical Database of the Czech Republic, an Interim Report. IANSA (Interdisciplinaria Archaeologica. Natural Sciences in Archaeology) 1: 49–53

Pokorný P (2002) A high-resolution record of Late-Glacial and Early-Holocene climatic and environmental change in the Czech Republic. Quaternary International 91:101–122

Pokorný P, Boenke N, Chytráček M, Nováková K, Sádlo J, Veselý J, Kuneš P, Jankovská V (2006) Insight into the environment of a pre-Roman Iron Age hillfort at Vladař, Czech Republic, using a multi-proxy approach. Veget Hist Archaeobot 15:419–433

Punt W (1976) The Northwest European Pollen Flora, Vol. I. Elsevier, Amsterdam

Punt W, Clarke GCS (1980) The Northwest European Pollen Flora, Vol. II. Elsevier, Amsterdam

Punt W, Clarke GCS (1981) The Northwest European Pollen Flora, Vol. III. Elsevier, Amsterdam

165 Punt W, Clarke GCS (1984) The Northwest European Pollen Flora, Vol. IV. Elsevier, Amsterdam

Punt W, Blackmore S, Clarke GCS (1988) The Northwest European Pollen Flora, Vol. V. Elsevier, Amsterdam

Punt W, Blackmore S (1991) The Northwest European Pollen Flora, Vol. VI. Elsevier, Amsterdam

Punt W, Blackmore S, Hoen PP (1995) The Northwest European Pollen Flora, Vol. VII. Elsevier, Amsterdam

Punt W, Blackmore S, Hoen PP, Stafford PJ (2003) The Northwest European Pollen Flora, Vol. VIII. Elsevier, Amsterdam

Punt W, Blackmore S, Hoen PP, Stafford PJ (2009) The Northwest European Pollen Flora, Vol. IX. Elsevier, Amsterdam

R Development Core Team (2010) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. http://www.R-project.org

Radu V (2005) Atlas for the identification of bony fish bones from archaeological sites. Editura Contrast, Bucuresti

Ramsey BC, Dee M, Lee S, Nakagawa T, Staff R (2010) Developments in the calibration and modelling of radiocarbon dates. Radiocarbon 523:953–961

Reimer PJ, Baillie MGL, Bard E, Bayliss A et al. (2009) IntCal09 and Marine09 Radiocarbon Age Calibration Curves, 0-50,000 Years Cal BP. Radiocarbon 514:1111–1150

Rösch M (1999) Ein Pollenprofil aus dem ehemaligen Fischweiher des Herzogs von Württemberg bei Nabern, Stadt Kirchheim/Teck, zur Kenntnis der Kulturlandschaftsgeschichte des Späten Mittelalters und der Frühen Neuzeit im Vorland der Schwäbischen Alb. In: Fundberichte aus Baden-Wurttemberg. 741–778, Konrad Theiss Verlag, Stuttgart

Rösch M (2012) Die Weiher des UNESCO-Welterbes Kloster Maulbronn als wirtschafts- und umweltgeschichtliche Archive. In Archologische Ausgrabungen in Baden-Wurttemberg. 71– 75, Konrad Theiss, Stuttgart

Rybníčková E, Rybníček K (1985) Paleogeobotanical Evaluation of the Holocene Profile from the Rezabinec Fish-Pond. Folia Geobotanica & Phytotaxonomica, 20(4):419-437

Sadori L, Mercuri AM, Mariotti M (2010) Reconstructing past cultural landscape and human impact using pollen and plant macroremains. Plant Biosystems 144 (4):940–951

Sasaki N, Takahara H (2011) Late Holocene human impact on the vegetation around Mizorogaike Pond in northern Kyoto Basin, Japan: a comparison of pollen and charcoal records with archaeological and historical data. Journal of Archaeological Science 38:1199–1208

166 Shannon CE (1948) A mathematical theory of communication. The Bell System Technical Journal 27:379–423 and 623–656

Schweingruber FH (1978) Mikroskopische Holtzanatomie. Kommissionsverlag Zürcher AG, Zug

Sládeček V (1986) Diatoms as indicators of organic pollution. Acta Hydrochimica et Hydrobiologica 145:555–566

Smol JP, Birks HJ, Last WM (2001) Tracking Environmental Change Using Lake Sediments. Volume 5: Data Handling and Numerical Techniques. Kluwer Academic Publishers, Dordrecht

Starec P (2005) Z příspěvků archeologie ke studiu vodních poměrů v historickém jádru Prahy [Archaeological contribution to the knowledge about watercourses in the historical centre of Prague]. Documenta Pragensia 24:39–46

Starec P, Houfková P, Bernardová A (2012) A contribution to the knowledge of natural conditions of Nové Město – excavations in Panská Street No. 1308/II. Staletá Praha 28:149– 164

Stockmarr J (1971) Tablets with spores used in absolute pollen analysis. Pollen et Spores 13:615-621

Święta-Musznicka J, Latałowa M, Badura M, Gołembnik A (2013) Combined pollen and macrofossil data as a source for reconstructing mosaic patterns of the early medieval urban habitats e a case study from Gdańsk, N. Poland. Journal of Archaeological Science 40:637-648

Ter Baraak C.J.F. & P. Šmilauer, 2002. Canoco Reference Manual and CanoDraw for Windows User’s Guide. Software for Canonical Community Ordination (version 4.5). Biometris, Wageningen, České Budějovice.

Théry-Parisot I., L. Chabal & J. Chrzavzez, 2010. Anthracology and taphonomy, from wood gathering to charcoal analysis. A review of the taphonomic processes modifying charcoal assemblages in archaeological contexts. Palaeogeography, Palaeoclimatology, Palaeoecology 291: 142–153.

Tomek V.V., 1892. Dějepis města Prahy I [History of the town of Prague]. Řivnáč, Praha. van Geel B., 2001. Non-pollen Palynomorphs. In Smol J.P. et al. (eds), Tracking Environmental Change Using Lake Sediments. Vol. 3: Terrestrial, Algal, and Siliceous Indicators, Kluwer Academic Publishers, Dordrecht, the Netherlands.

Wheeler A. & A.K.G. Jones, 2009. Fisches. Cambridge University Press, Cambridge.

Zavřel J., 2006. Geologická problematika archeologického poznání Nového Města pražského [Contribution of geology to the archaeological knowledge of the New Town of Prague]. Archaeologica Pragensia 18: 245–261

167

168

Paper III

Archaeobotanical Database of the Czech Republic, an Interim Report

AdélaPokorná, Dagmar Dreslerová, Dana Křívánková

169 Archaeobotanical Database of the Czech Republic, an Interim Report

AdélaPokorná, Dagmar Dreslerová, Dana Křívánková

Abstract

Archaeobotanical research in the Czech Republic has generated enormous quantities of unique, but non-uniform, data that are currently stored in disparate hard copy and electronic formats. A consortium of archaeological organisations has initiated the Archaeobotanical Database of the Czech Republic (CZAD) to facilitate data storage, access and quality. CZAD is being developed under the auspices of the ArboDat Project, by the Institute of Archaeology, Prague, and Landesamt für Denkmalpflege, Hessen, Ger. CZAD is based on the international ArboDat Multi version of the ArboDat database originally developed in Germany. Adaptations have been made to suit local requirements. As at the end of 2010 more than half of the existing data in the Czech Republic has been entered into CZAD. Data entry is on a strictly voluntary basis and author’s data and copyright are protected.

Introduction

The analysis of plant macro-remains is a useful tool for obtaining archaeobotanical data that can help solve many issues, including such diverse questions as: past subsistence strategies; crop husbandry regimes; human diet; vegetation and climate changes; or effects of human activities on the landscape. It can further our understanding of the use of wild plants for medicinal and technical purposes and help us, for example, to reconstruct long distance trade in exotic crops. Today, archaeobotany’s position as a respected branch of science has arisen both from a gradual awareness of the high information content of biological data from archaeological investigations, and an increasing interest in the everyday life of past humans and their relationship with the local environment.

The increasing number of macro-remains analyses in recent decades has led to the urgent need for a uniform treatment of data in an electronic form. To this end the new Archaeobotanical Database for the Czech Republic (CZAD) was developed at the Institute of Archaeology, Prague. In this article we outline the factors that lead to the creation of the database, of its philosophy and technical execution, and the current state of its development.

Computing archaeobotanical data

The idea to design a new database for the rapidly accumulating archaeobotanical data arose several years ago. Researchers from the Laboratory of Archaeobotany and Palaeoecology (LAPE); the West Bohemian Institute for Protection and Documentation of Monuments (ZIP); and the Institute of Archaeology in Prague (ARUP) agreed to use a unified database that allowed for the efficient exchange, processing and utilization of data. The database would also make possible the centralization and archiving of all archaeobotanical data in the Czech Republic. However, the creation of a new database system might lead to complications in subsequent steps in data synthesis, such as the interpretation of archaeobotanical data from Czech localities within a larger European context.

170 The decision was therefore taken to use an existing database program ArboDat developed by A. Kreuz and E. Schäfer of the Landesamt für Denkmalpflege Hessen in Wiesbaden, Germany. This program has very sophisticated utilities for archiving and evaluating archaeobotanical data (Kreuz, Schäfer 2002). It was created using MS:Access, designed and adapted for handling archaeobotanical data, with specific adaptations allowing supra-regional and international collaboration on questions that go beyond the borders of individual states. In addition to its archival function, the program incorporates tools for data evaluation and further processing. The hierarchical organization of ArboDat enables easy access not only for archaeobotanical data, but also for background archaeological and ecological information concerning sites, archaeological features, archaeobotanical samples and the ecological characteristics of plant species. ArboDat is already being used by a range of archaeobotanists, including many in Germany, Austria, Switzerland, Denmark, the Netherlands, Belgium, France, Poland and Finland.

In 2009, ARUP established a cooperation with Landesamt für Denkmalpflege Hessen in Wiesbaden (represented by Prof. Angela Kreuz). The bilateral ArboDat project, supported by the Academy of Science of the Czech Republic (ASCR) was launched for the purpose of:

1) translating the original German version of the program into English to enable more widespread use of the software in non-German-speaking countries. This was developed later into a multilingual version of ArboDat, called ArboDat Multi (software solution D. Křivánková), which is based on a universal program code. The universal code enables the use of ArboDat Multi in various language versions and still allows universal upgrades of the program independent of the language. Any new language version of the database can be made by simply translating and filling in all the necessary terms into the translation tables that are an inseparable part of the database. In cooperation with A. Kreuz and the team of S. Thiebault (Department Archéozoologie-Archéobotanique, CNRS, Paris, France), three language versions of ArboDat Multi have so far been prepared (Czech, French and English) in addition to the original German. The local or ‘country’ versions of the ArboDat cater to the specific needs of each country (e.g. the listing of archaeological cultures or phyto-geographical areas). Three local versions now exist for Germany, France and the Czech Republic. 2) establishing and preparing CZAD using ArboDat allow: (a) the collection of data (results of archaeobotanical analyses both published and unpublished); (b) online access for the professional community to basic information in CZAD; (c) a facility to exchange and share data between archaeobotanists, both in the Czech Republic and abroad; and (d) an overview of all archaeobotanical data that will help in identifying preferences for further investigation. There is general agreement that CZAD data will be stored at ARUP. This arrangement will allow interconnection between CZAD and the Czech Archaeological Database which would substantially improve the objective of combining both archaeological and archaeobotanical data. ARUP guarantees protection of authors’ copyright and data. The provision of data to CZAD will be on a voluntary basis.

ArboDat program

ArboDat and its adaptations enable the insertion and further processing of results from macro- remains analyses (including wood and charcoal). The uniform coding of plant taxa is fundamental to the subsequent exchange of the data. A system created by S. Jacomet in Basel (Paulsen 1995)

171 has been used as the basis for the plant coding (PCODE). The results section of ArboDat contains, among other things, the PCODE of each identified macro-remain, the type of macro-remain (i.e. seed/fruit, rachis, root, wood, etc.), the state of its preservation (i.e. charred, desiccated, etc.) and the number of macro-remains of the taxon (this counting is subject to detailed rules listed in the manual of the program).

Figure 1. Hierarchical organization of the ArboDat database (Kreuz, Schäfer 2002): Form view of the input mask. Information concerning the archaeological site and the context of analysed samples are structured hierarchically (see Form view in Figure 1). Some selected data are entered through pull- down menus, in particular data essential for further evaluation of results (archaeological dating, feature type, site type, type of sampling, type of macro-remain, preservation of macro-remain, etc.) The structure of the database also includes ecological information concerning the plant taxa involved (for more detailed information see Kreuz, Schäfer 2002). ArboDat also contains a dozen pre-programmed queries designed for the evaluation of data according to the choice and requirements of the user.

As the ArboDat was originally designed for use in German-speaking countries, all the forms, tables and structural data were obviously written in the German language. Moreover, the whole database system had been designed according to the requirements of German users. Therefore, some changes in the structural data (e.g. the list of archaeological cultural groups or phyto-geographical units) had to be made and the appearance of some forms adapted, for the program to be used in our country. These changes were incorporated into the structure of the multilingual version (ArboDat Multi, see above).

Table 1. Quantitative overview of the archaeobotanical data inserted into the database up to the end of 2010 - according to archaeological periods. Only selected cultures are mentioned.

Period/culture Sum of projects Sum of samples Macro-remains

172 Mesolithic 1 17 82

Neolithic 16 115 3356

Linear Pottery Culture 4 36 1106 Stroked Pottery Culture 4 10 205 Lengyel Culture 1 52 403

Eneolithic 18 193 2814 Corded Ware Culture 3 36 143 Jordanov Culture 2 46 436 Bell Beaker Culture 2 48 220

Bronze Age 39 957 35917

Únětice Culture 3 68 946 Knovíz Culture 5 82 6182 Lusatian Culture 2 35 858 Štítary Culture 1 26 5567

Early Iron Age 15 205 24054

Platěnice Culture 1 23 3180 Hallstatt period 14 182 20874

La Tène Period 10 224 3696

Roman Period 4 66 5542

Migration Period 1 14 293

Medieval Period 121 1165 635570

Early Middle Ages 26 166 202012 High Middle Ages 70 533 330585

Post-medieval Period 25 89 165235

CZAD database

173 Archaeobotanical identification of plant macro remains from prehistoric and medieval archaeological contexts has an eighty years tradition in the Czech Republic (Čulíková 2004; Dreslerová 2008; Kočár, Dreslerová 2010). The vast amount of archaeobotanical data from nearly 300 prehistoric and several hundred medieval sites has led us to create a centralized database to incorporate all archaeobotanical data from our country. CZAD is open to all data from archaeological and off-site contexts. Moreover, the database program is also designed for antracological data and results of wood fragment identification.

Two types of data sources are available – in written and electronic form. Data existing in written form cannot be entered into the database other than manually.. However, a special macro is being developed for electronic data to transform MS:Excel tables directly to the MS:Access tables used in ArboDat. Data being currently incorporated into CZAD includes: (a) reports in the ARUP archive; (b) reports provided by individual authors; and (c) published data.

Up to the end of 2010 carpological data from 180 selected archaeological sites had been entered into the database (the only motive for selection was the availability of data). This is roughly more than half of the sites investigated archaeobotanically.

Entered data represent some 3770 processed samples, while the total amount of macro- remains involved exceeds almost one million. For a general overview of the data that has been already incorporated into the database see Table 1. The Medieval Period is represented by the majority of sites (121). In contrast, both the Mesolithic and Roman Period are underrepresented in our data (1 site for each period). The spatial distribution of sites with archaeobotanical finds within the middle European region is shown in Figure 2 - a) and b). A large proportion of the sites are concentrated near cities, a consequence of the increased building activity found there.

It should be noted that all information indicated in both maps on Figure 2 and Table 1 are still interim results; as the input of data to the database is an ongoing process. 5 This is the main reason why we have avoided presenting any archaeobotanical data in this text. However, for a small example of the possibilities of the database and the further use of its stored data see Figure 3.

CZAD was created with the intention of ensuring continuity between earlier analyses, those now in progress, and those done in the future. A centralised database has great advantages for the maintenance and use of all data, particularly in the case of unexpected occurrences (e.g. an author leaves the professional field, an archaeological or archaeobotanical unit ceases to exist, etc.).

5At the time of proofreading of this article the number of archaeological sites inserted to the database increased to 265. The list of the authors of the analyses incorporated by now is as follows: E. Opravil, V. Čulíková, Z. Tempír, P. Kočár, V. Komárková, A. Bernardová and A. Pokorná.

174

Figure 2. Spatial distribution of archaeological sites with archaeobotanical data within the Czech Republic (status as of end of 2010). (a) Prehistory: black – data already inserted into the database; red – data not yet inserted (according to Kočár, Dreslerová 2010). (b) Early to Post Medieval data inserted into the database.

Thanks to the current project’s funding there is the possibility to enter the bulk of the published data and unpublished reports from individual authors’ archives into CZAD up to the end of 2011. In subsequent years only data that is processed by individual authors in the ArboDat database format will be received and inputed. Updating of the centralised database and merging new data will be performed once a year, in a manner similar to that used by CNRS, France).

The specific rules regarding accessibility, handling and publication of data will be based on the requirements of data contributors. The upgraded version of the merged database will be available to all contributors on CD. ARUP guarantees the protection of authors’ copyright and data. The actualised list of localities incorporated into CZAD will be accessible on the CZAD web page from the end of 2011.

70 Figure 3. Illustrative example of one use of the archaeobotanical database: weeds and ruderal plants encountered 60 in assemblages of macro-remains in the Czech Republic. Numbers of taxa 50 according to individual archaeological periods (correct up to the end of 2010). 40 3 Periods: NE - Neolithic, EN - 4 Eneolithic, BR - Bronze Age, HA - 6 30 7 Hallstatt, LA - La Tène, MA - Medieval. Ecological groups: 3 -

20 ruderal vegetation, 4 - ruderal/segetal vegetation, 6 - weeds of root-crops and gardens, 7 - weeds of cereals. The 10 classification of taxa follows the individual ‘ecological’ grouping as 0 given by the index in the ArboDat NE EN BR HA LA MA program (Kreuz, Schäfer 2002; Kreuz, Schäfer 2010).

175

Concluding remarks

Plant macro-remains analysis produces a vast amount of archaeobotanical data, of differing information value in relation to their quantity, site location, time period or excavation and processing methods. A centralized archaeobotanical database may increase the relevance of each of the analyses included: even seemingly marginal or less numerous findings can help to complete a mosaic of discontinuous data and enable a better understanding of the questions asked.

We believe that CZAD will significantly improve archaeobotanical research in our country. With its help better results may be achieved in a number of research areas, such as the first intentional production of crops, regional and supra- regional agricultural practices, or the migration of plant species, to name but a few. Moreover, a unified database system allows specialists to cooperate at an international level. We hope that the many contributors will participate in its continuing development; despite the fact that the provision and sharing of data is on a strictly voluntary basis, this integrated database is intended for the general benefit of all.

Acknowledgements The work on CZAD was made possible thanks to financial support under a research grant of the Academy of Sciences of the Czech Republic (AS CR) M300020902. We would like to thank Steve Ridgill for improving our English. We would also like to thank Prof. Angela Kreuz for fruitful collaboration.

References

Čulíková V (2004) Archeobotanika v české archeologii na prahu 3. tisíciletí. Archeologické rozhledy 56:661–671

Dreslerová D (2008) Better later than never: environmental archaeology in the Czech Republic. In: Beneš, J., Pokorný, P. (eds) Bioarchaeology in the Czech Republic. Jihočeská univerzita v Českých Budějovicích, Přírodovědecká fakulta, Archeologický ústav Akademie věd České republiky, Praha, České Budějovice 13–38

Kočár P, Dreslerová D (2010) Archeobotanické nálezy pěstovaných rostlin v pravěku České republiky. Památky archeologické 101:203–242

Kreuz A, Schäfer E (2002) A new archaeobotanical database program. Veget Hist Archaeobot 11:177–179

Kreuz A, Schäfer E (2010) Archäobotanisches Datenbankprogramm ArboDat – Handbuch. Landesamt für Denkmalpflege Hessen, Wiesbaden

Kuneš P, Abraham V, Kovařík O, Kopecký M & PALYCZ contributors (2009) Czech Quaternary Palynological Database PALYCZ: Review and basic statistics of the data. Preslia Praha 81:209–238

Paulsen J (1995) Namen und Synonyme mitteleuropäischer Gefässpflanzen. Schweizer Botanik CD 1995. Botanisches Institut der Universität Basel, Basel

176

177

Paper IV

Ancient and early medieval human–made habitats of the Czech Republic: Colonization history and vegetation changes

Adéla Pokorná, Petr Kočár, Jiří Sádlo, Tereza Šálková, Pavla Žáčková, Veronika Komárková, Zdeněk Vaněček, Jan Novák

178 Ancient and early medieval human–made habitats of the Czech Republic:

Colonization history and vegetation changes

Pokorná A, Kočár P, Sádlo J, Šálková T, Žáčková P, Komárková V, Vaněček Z, Novák J

Introduction

In Europe, the vascular plants of alien origin are divided into two groups according to their residence time. The archaeophytes were introduced, either intentionally or unintentionally, during the time span between the Neolithic and the High Middle Ages, whereas the neophytes have immigrated since the beginning of the Modern period (e.g. LaSorte et Pyšek 2009). The neophytes are generally studied both on a grand scale and in detail, and their familiarity provokes their further research. Contrarily, the archaeophytes are rather neglected in this respect, though they are often used as a reference group in relation to the neophytes.

Nevertheless, the archaeophytic flora is numerous (ca 300 species in central Europe) and the archaeophytes differ conspicuously in their residence time, distribution, and invasive success as well as in their conservation status, which makes them to be attractive objects of investigation. The progressive immigration of archaeophytes lasted for more than a half of the Holocene period. Thus, conditions of species immigration developed dramatically during this period, and archaeophytes appear a complex and diversified chronological group of plants needing further study in this respect.

The delimitation of the archaeophytic stage is rather robust and it is connected with a huge leap in the development of cultural landscape in Europe. Its start (ca 5500 BC, halfway of Holocene climatic optimum; Kalis 2003) is associated with the beginning of agricultural activities, which comprised the introduction of alien crops as well as weeds, the beginning of sedentary life, and also the gradual emerging of the cultural landscape (Pokorný et al. 2015). The end of the archaeophytic stage, at about 1500 AD, is often labelled in the botanical literature as a time when America was discovered, which is, however, a rather simplified interpretation. The onset of the modern times, taking place at the beginning of cool oscillation of the so called Little Ice Age, is defined by a substantial cultural and economic change. It was given, besides other things, by consolidations and further expansions of several large political units, such as Spanish, Roman and Ottoman Empires, as well as the Grand Duchy of Moscow, and by the early transmarine expeditions all around the world.

Two different approaches are valid in understanding the long-term developmental changes of plant cover in the past. The first approach, called secular succession (van der Maarel 1988, or synchronology sensu Braun-Blanquet 1964), describes the development of plant cover as a long–term vegetation process driven by both the climate and cultural influences, such as lifestyle and management practices. The second approach, called colonization history, describes this process in a perspective of the ecology of invasions. This approach can be applied both to the newly established alien plants, e.g. archaeophytes, and to those native ones, which originate from the surrounding landscape and locally colonize anthropogenic habitats. The concept of

179 our work is closer to the latter approach, handling with the particular data without aspiration to a generalized holistic description.

In the colonization history of aliens, the residence time is a crucial attribute, expressing the time span since a particular species introduction (Wilson et al. 2007). The size of a species range, the frequency of its current distribution, as well as its invasion status depends on this particular parameter (for more details see Pyšek–Jarošík 2005, Richardson–Pyšek 2006). Since it is usually not known exactly when a taxon was introduced, we adopt the term ‘Minimum Residence Time’ (MRT) expressing a time span since the first record of a species in the country (Rejmánek 2000, Pyšek et al. 2015).

The impact of colonization history on invasion success of a species is frequently studied in neophytes (e.g. Pyšek–Jarošík 2005), because reliable information of their residence time is mostly accessible, and the whole process since its introduction to its invasion can be easily traced. In the case of archeophytes, however, the attempts to historical categorization at a regional scale are still rare (e.g. Preston et al. 2004, Brun 2009).

We tried to fill this gap in the knowledge of aliens’ history. We gained macroremain material from archaeological settlement sites. It allowed us an appropriate dating, however, owing to taphonomical problems (Greig 1981, Behre–Jacomet 1991, Heimdahl 2005, Bosi et al. 2011, Święta-Musznicka et al 2013), our data set necessarily represents only a certain portion of the original local vegetation context and therefore it cannot be entirely released from imperfections. To picture the vegetation context of the past settlements and adjacent landscape, we studied both the aliens and the native species. According to established MRT of individual species we inferred regional developmental scenario and periodization of the archeophytic stage.

We work with concept of ancient cultural landscape which is broadly shared in Central European literature, being inferred from paleodata analyses (Pokorný et al. 2015) as well as modern analogies (Dreslerová–Sádlo 2000). A considerable decrease of human impact is assumed between settlements and surrounding landscape in which ruderal sites and fields are considered typical human-made habitats. However, definition of the opposite extreme is fuzzy owing to semi-natural grasslands which may or must not be included. Only outside the past cultural enclaves, the vegetation preserved a rather natural character (e.g. forests) despite its possible human exploitation.

Aims

Our goal is to understand remote developmental phases of synanthropic flora of the Czech Republic, specifically period since the early Neolithic to the end of the Early Middle Ages. We address the following aims: (i) to accomplish quantitative characterization representing developmental changes of species diverzity both in alien and native species; (ii) to attempt estimantion of past vegetation development using species groups defined by their common residence time and recent habitat link; (iii) in alien species, to ascertain impact of their current abundance and invasiveness to their residence time.

180 Materials and methods

Data set

Macro–remain assemblages originated in archeological contexts were gained from (i) Archaeobotanical Database of the Czech Republic (Pokorná et al 2011, Dreslerová et al. 2015, http://www.arup.cas.cz/czad/?l=en), and (ii) unpublished data of the authors of this study. All data were primarily processed using the archaeobotanical database programme ArboDatMulti (Pokorná et al. 2011, Kreuz–Schäfer 2002).

Our data contained ca. 5,000 records of more than two hundred plant taxa from 223 sites. Each site represents one cultural phase of 197 localities dated to the time span between the early Neolithic (5600 BC) and the end of the Early Medieval Period (1200 AD). The data from the High Middle Ages (ca. 1200 to 1500 AD) couldn’t be integrated to the study, because the process of filling data to the database is still under the process. But important is that they are hardly comparable to ancient cultures owing to their ample amount and different taphonomy. Most of the analysed material has been preserved by charring, with several exceptions of waterlogged sites. The majority of sites were analysed between the years 2000 and 2014 (the oldest analysis comes under ca. 1960) and about one half of the reports haven’t been published yet in the scientific journals. The number of independent researchers was 12.

Plant species and morphotaxa

Visualy-based taxonomic identification of macro-remains resulted in the list of morphotaxa. Since most of them were identified at the species level, we generally use the term species instead of morphotaxon. Taxa representing more than one species came either from original reports of the database or were newly established to minimize potential identification bias adopted from various database sources. Following collective taxa were used: Ajuga genevensis/reptans; Amaranthus blitum/graecizans; Arctium cf. lappa/tomentosum; Chenopodium glaucum/rubrum; Galeopsis angustifolia/ladanum; Lamium amplexicaule/purpureum; Melilotus albus/officinalis; Papaver dubium/rhoeas; Potentilla sp.; Rumex crispus/obtusifolius; Setaria verticillata/viridis; Stachys annua/arvensis; Veronica polita/opaca; Taraxacum sp.; Vicia cracca/sepium; Vicia hirsuta/tetrasperma; Vicia pannonica/sativa. According to their immigration status, adopted from the list of Danihelka et al (2012), which is valid for the Czech Republic, the species were assorted into two groups: (i) natives (i.e. indigenous) and (ii) aliens (i.e. introduced, exotic, adventive).

We targeted to herbs which presumably grew spontaneously in settlements and entered the fossil record unintentionally. Therefore we excluded the following groups from our primary data: (i) trees and shrubs, (ii) species of natural habitats such as rocks or water pools , and (iii) crops harvested intentionally (e.g. Triticum sp., Pisum sativum), as well as wild herbs collected presumably as a food (e.g. Fragaria sp.) or for medical purposes (e.g. Atropa bella–dona). Contrarily, we included several species of open land although they produce edible seeds (e.g. the genera of Bromus, Chenopodium, Fallopia, Sambucus, Stipa and Vicia), which may have been semi–spontaneous, i.e. both wild and perhaps also potentially supported by human intention (see e.g. Bieniek 2002). Owing to the methodical reasons, we further excluded also

181 (iv) the collective taxa in which both native and alien species were combined, (v) rare native species occurring in less than 5 sites during all periods, and (vi) species with extremely discontinuous occurrence in our data, in which, therefore, it was not possible to specify the MRT.

Tab. 1 Chronology of examined data. Absolute dating of prehistoric periods for the Czech Republic follows Jiráň– Venclová (2013). Sum_seed means total number of determined seeds of studied species.

Length Number of Abbr. Periods of the Prehistory Time span (years) sites Sum_seed LBK Early Neolithic (Linear Pottery c.) 5600 – 4900 BC 600 10 5894 NEO Late Neolithic 5000 – 4200 BC 700 11 2182 ENE Eneolithic 4500 / 4400 – 2300 BC 2000 17 1107 BR1 Early to Middle Bronze Age 2300/2200 – 1250 BC 1000 20 4793 BR2 Late to Final Bronze Age 1250 – 800/750 BC 500 46 20793 EIA Early Iron Age (Hallstatt) 800 – 400/370 BC 350 30 17467 LIA Late Iron Age (La Tène) 480 / 460 – 50 / 20 BC 450 22 9653 RMP Roman to Migration Period 35 / 25 BC – 560 / 580 AD 580 16 9926 EM1 Early Middle Ages 1-3 580 – 950 AD 370 15 71824 EM2 Early Middle Ages 4 950 – 1200 AD 250 36 76534

Sites and chronology

First of all we excluded from the database the sites containing less than 5 species corresponding to the above criteria, and the sites suspected from material contamination, mistakes in plant determination or doubtful archaeological dating. To reflect the differences in cultural/socioeconomic development among individual periods, we chose classification into ten chronological phases summarized in the Tab. 1. For general information about the localities see Tab. 2.

The Neolithic period (Pavlů–Zápotocká 2007) was divided into two phases: the initial early Neolithic or Linear Pottery Culture (LBK) and successive late Neolithic (LNE), owing to distinguish the earliest wave of invasion of the alien plants. The Eneolithic period was used in its usual definition (Neustupný 2008). In contrast to the commonly used archaeological periodization of the Bronze Age (Jiráň 2008), we divided it into two phases: the early to middle Bronze Age (BR1) and the late to final Bronze Age (BR2), differing considerably from each other in both the technological progress in agriculture and the range of cultivated crops (Dreslerová–Kočár 2013, Dreslerová et al. 2017). Subsequent Early Iron Age (EIA) and Late Iron Age (LIA) are used in common sense of the Hallstatt and the La Tène cultures, respectively (see Venclová 2008a,b). Contrarily, the short Migration Period was attached to the previous Roman Period (RMP), because in the Czech territory both periods are close to each other from both the cultural and technological point of view (Jiráň & Venclová 2013). Early Medieval was divided into two periods (EM1 and EM2) reflecting population increase and political changes, as well as the homogeneity of the archaeobotanical data available.

182 Tab. 2 List of sites according to periods.

land registry references pres.

LBK BYLANY U KUTNÉ HORY Rulf–Tempír 2002 c HULÍN Kočár, Vaněček unpubl. c MOHELNICE Tempír 1968, Opravil 1979 w OPAVA - MĚSTO Tempír unpubl. c PRAHA - DEJVICE Žáčková unpubl. c PRAHA - HLUBOČEPY Kočár unpubl. c PRAVČICE Kalábek et al. 2010 c PŘEŠTICE Kočár unpubl. c TACHLOVICE Kočár unpubl. c ÚSTÍ NAD LABEM Kočár unpubl. c LNE HULÍN Kočár, Vaněček unpubl. c KOLÍN Kočár–Kočárová 2012 c KOTOPEKY Kočár unpubl. c LOUČKA Komárková unpubl. c NEMILANY Vaněček unpubl. c PRAHA - DEJVICE Žáčková unpubl. c PRAVČICE Kalábek et al. 2010 c PŘÍŠOVICE Kočár unpubl. c TACHLOVICE Kočár unpubl. c TĚŠETICE U ZNOJMA Šabatová et al. 2012 c ENE BOŘITOV Kühn 1981a c HULÍN Kočár, Vaněček unpubl. c KROMĚŘÍŽ Berkovec et al. 2005 c KUTNÁ HORA Čulíková 2009 c MOHELNICE Tempír 1968, Opravil 1979 w OPAVA - MĚSTO Kočár unpubl. c PRAHA - LIBOC Kočár unpubl. c PRAHA - MICHLE Kočár unpubl. c PRAHA - MIŠKOVICE Ernée et al. 2007 c PRAHA - RUZYNĚ Žáčková unpubl. c PRAVČICE Kalábek et al. 2010 c SLAVONÍN Kočár unpubl. c STARÝ PLZENEC Komárková unpubl. c STAVENICE Vaněček unpubl. c TUCHOMĚŘICE Komárková unpubl. c VITČICE NA MORAVĚ Kočár unpubl. c ŽELEČ U ŽATCE Kočár unpubl. c BR1 BRANDÝS NAD LABEM Danielisová et al. 2013 c DOBŘANY Kočár unpubl. c HULÍN Kočár, Vaněček unpubl. c JAVORNICE U DUBU Šálková unpubl. c KOLÍN Kočár & Kočárová 2012 c MEDLOV U UNIČOVA Kočár unpubl. c MORAVSKÝ KRUMLOV Klečka–Skutil 1937, Tempír 1966, Kühn 1981b c NEMILANY Vaněček unpubl. c PAVLOV U DOLNÍCH VĚSTONIC Opravil 2002 c PEČKY Kočár unpubl. c PLANÁ U ČESKÝCH BUDĚJOVIC Šálková unpubl. c PRASKLICE Tempír 1961, 1966, 1968, Kühn 1981b c PRAVČICE Kalábek et al. 2010 c PRAVČICE Vaněček unpubl. c ŠLAPANICE U BRNA Kühn 1981c c TĚŠETICE U ZNOJMA Šabatová et al. 2012 c VITČICE NA MORAVĚ Kočár unpubl. c VLINĚVES Kočár unpubl. c VRCHOSLAVICE Kočár unpubl. c VRCOVICE Hlásek et al. 2014 c BR2 BAVORYNĚ Kočár unpubl. c BDENĚVES Komárková unpubl. c BLUČINA Salaš et al. 2012 c BRANDÝS NAD LABEM Danielisová et al. 2013 c BŘEZNICE U BECHYNĚ Šálková unpubl. c ČERNOŠICE Tempír 1985 c

183 land registry references pres. ČERNÝŠOVICE Šálková unpubl. c DRAHELČICE Kočár unpubl. c DUBICKO Vaněček unpubl. c HOSTIVICE Komárková unpubl. c HVOŽĎANY U BECHYNĚ Chvojka et al. 2011 c JÍVOVÁ Kühn 1981b c KLATOVY Kočár unpubl. c KNĚŽEVES U PRAHY Kočár–Mihályiová 2011 c KNĚŽÍVKA Kočár unpubl. c KOLÍN Kočár–Kočárová 2012 c LOŠTICE Nekvasil–Opravil 1994 c LOVOSICE Čulíková 2008 c MEDLOV U UNIČOVA Kočár unpubl. c MĚRUNICE Kočár unpubl. c OBORY Šálková unpubl. c OSTROV U STŘÍBRA Kočár unpubl. c PEČKY Kočár unpubl. c PÍSEK Jiřík et al. 2012 c PRAHA - BUBENEČ Kočár unpubl. c PRAHA - DOLNÍ CHABRY Kočár unpubl. c PRAHA - HOSTIVAŘ Šmejda–Kočár 2007 c PRAHA - LIBOC Kočár unpubl. c PRAHA - RUZYNĚ Žáčková unpubl. c PRAHA - SMÍCHOV Kočár unpubl. c PRAHA - ZÁBĚHLICE Kočár unpubl. c PŘEŠTICE Kočár unpubl. c ROZTOKY U PRAHY Tempír 2012 c SEZEMICE NAD LOUČNOU Kočár unpubl. c SLANÝ Vaněček unpubl. c ŠLAPANICE U BRNA Kühn 1981c c SLAVONÍN Kočár unpubl. c SOBĚTICE U KLATOV Kočár unpubl. c TUCHOMĚŘICE Kočár–Kočárová 2007 c TURNOV Komárková unpubl. c VELKÉ PŘÍLEPY Komárková unpubl. c VSTIŠ Kočár unpubl. c ZDICE Kočár unpubl. c ZHOŘ U TÁBORA Chvojka et al. 2014 c EIA BAŠŤ Kočár unpubl. c BDENĚVES Kočár unpubl. c BDENĚVES Komárková unpubl. c HABRŮVKA Fietz 1941, Kühn 1972, Tempír 1961, 1968 c HRUŠOV NAD JIZEROU Kočár unpubl. c HULÍN Kočár, Vaněček unpubl. c KAL U KLATOV Kočár unpubl. c KLIMKOVICE Komárková unpubl. c KROMĚŘÍŽ Berkovec et al. 2005 c LAŽÍNKY Kočár unpubl. c MEDLOV U UNIČOVA Kočár unpubl. c MĚRUNICE Kočár unpubl. c MĚSTO BRNO Kočár unpubl. c NEMILANY Vaněček unpubl. c PRAHA - ČERNÝ MOST Vaněček unpubl. c PRAHA - HLOUBĚTÍN Kočár unpubl. c PRAHA - HOSTIVAŘ Kočár unpubl. c PRAHA - LIBOC Kočár unpubl. c PRAVČICE Kalábek et al. 2010 c PŘEŠTICE Kočár unpubl. c RAJHRAD Kühn 1980 c SOBĚTICE U KLATOV Kočár unpubl. c TACHLOVICE Kočár unpubl. c TĚŠETICE U ZNOJMA Šabatová et al. 2012 c ÚVALNO Kühn 1960, 1981b, Tempír 1966, 1968 c LIA HNĚVOTÍN Vaněček unpubl. c HOLUBICE V ČECHÁCH Kočár unpubl. c HRUBÁ VRBKA Kočár unpubl. c HULÍN Kočár, Vaněček unpubl. c JINOČANY Kočár unpubl. c KOLÍN Kočár–Kočárová 2012 c LOVOSICE Čulíková 2008 c MALÉ HRADISKO Kühn 1981b c

184 land registry references pres. MĚSTO BRNO Kočár unpubl. c MŠECKÉ ŽEHROVICE Opravil 1998 c OKNA V PODBEZDĚZÍ Dreslerová et al. 2013 c OPLOT Komárková unpubl. c OSEK U MILEVSKA Fröhlich 2004 c PRAHA - DEJVICE Žáčková unpubl. c PRAHA - JINONICE Kočár unpubl. c PRAHA - KBELY Kočár unpubl. c PRAHA - VYSOČANY Kočár unpubl. c ŘEPČÍN Kalábek–Kočár 2007 c ZÁHOŘICE Pokorný et al. 2006 w ZÁHOŘICE Kočár unpubl. w ZÁHOŘICE Chytráček a kol. 2010, Chytráček et al. 2012, Šmejda et al. 2010 w, c RMP BŘEZNO U LOUN Tempír 1982 c DRNHOLEC Opravil 1989, 2002 c DROUŽKOVICE Kočár unpubl. c HOLUBICE V ČECHÁCH Kočár unpubl. c HRUBÁ VRBKA Kočár unpubl. c PASOHLÁVKY Kočár unpubl. c PRAHA - DEJVICE Žáčková unpubl. c PRAHA - HLOUBĚTÍN Kočár unpubl. c PRAHA - KBELY Kočár unpubl. c PRAHA - KŘESLICE Komárková unpubl. c PŘEŠŤOVICE Šálková unpubl. c RATAJE U BECHYNĚ Chvojka et al. 2011 c ROZTOKY U PRAHY Komárková unpubl. c VRCHOSLAVICE Kočár unpubl. c EM1 MIKULČICE Opravil 1998 w PRAHA - HLOUBĚTÍN Kočár unpubl. w PRAHA - HRADČANY Čulíková 2001 w PRAHA - MALÁ STRANA Kočár unpubl. w PRAHA - MALÁ STRANA Čulíková 1998a w PRAHA - MALÁ STRANA Čulíková 2005 w PRAHA - MALÁ STRANA Čulíková 2001 w EM2 HRADEC NAD MORAVICÍ Opravil 1992 c CHEB Komárková unpubl. w CHRUDIM Kočár unpubl. w, c CHRUDIM Komárková unpubl. w LIBICE NAD CIDLINOU Kozáková et al. 2014 w MĚSTO BRNO Kočár unpubl. w MIKULČICE Opravil 1998 w MIKULČICE Opravil 1972 w MOST Čulíková 1986 w OLOMOUC - MĚSTO Opravil 1985 w OLOMOUC - MĚSTO Opravil 1994 w PRAHA - HRADČANY Čulíková 1998 w PRAHA - HRADČANY Čulíková 2001 w PRAHA - HRADČANY Čulíková 2001 w PRAHA - HRADČANY Čulíková 2001 w PRAHA - HRADČANY Komárková unpubl. w, c PRAHA - MALÁ STRANA Opravil 1986 w PRAHA - MALÁ STRANA Kočár unpubl. w PRAHA - MALÁ STRANA Čulíková 1998a w PRAHA - MALÁ STRANA Čulíková 2005 w PRAHA - MALÁ STRANA Čulíková 2010 w PRAHA - MALÁ STRANA Čulíková 2001 w PRAHA - MALÁ STRANA Čulíková 2001a w PRAHA - NOVÉ MĚSTO Pokorná et al. 2014 w PRAHA - STARÉ MĚSTO Opravil 1986 w PRAHA - STARÉ MĚSTO Kočár unpubl. w, c PRAHA - STARÉ MĚSTO Komárková unpubl. w STARÁ BOLESLAV Kozáková et al. 2014 w STARÉ MĚSTO U UHERSKÉHO HRAD. Opravil 1980 w UHERSKÝ BROD Opravil 1976 w ÚSTÍ NAD LABEM Kočár unpubl. w

185 Assessment of temporal changes

In a quest of simple expression, following variables were plotted against ten subsequent periods situated at time axis: (i) cummulative and (ii) absolute numbers of newly occurring species, and (iii) percentual proportions of aliens and natives in each of individual periods. Principle component analysis (PCA) was performed using Canoco (Lepš–Šmilauer 2003) based on total floristic composition of individual periods.

The impact of residence time of aliens to their current invasiveness was studied using scale quantifying contemporary invasion success of archaeophytes. Cathegories of (i) invasive, (ii) common, (iii) scaterred, (iv) rare and (iv) threatened/extinct species were classificated. Group of invasive species correspond to Catalogue of Pyšek et al (2012). Common and scattered species were defined using recent floristic database within PLADIAS - integrated database on Czech plant diversity (www.pladias.ibot.cas.cz) and distinguished mutually using boundary value of 3000 recent findings. Red list (Grulich 2012) was used for defining of the latter two groups where joining of C3 and C4 cathegories define group of rare species, and threatened/extinct species were defined by merging of other ones (A1, A2, C1 and C2).

Floristic development based on MRTs and ecological demands of species

MRT values were defined for both alien and native species. In the former group, this trait refers to the possible time of their immigration at regional scale whereas in the latter it means rather the age of their first local occurrence in fossil seed material. Then we classified the species into groups according to their similar MRTs and their present ecological demands, derived from regional data (Chytrý et al. 2009 for recent habitats and vegetation units; and unpublished data intended for Pladias database). Four phases were distinguished within the archaeophytic stage, each phase characterized by a species group, which then markedly proved for the first time. Some of the species groups are described using phytosociological units (names were adopted from Chytrý et al. 2009).

Results

Immigration dynamics

Among 217 taxa meeting criteria, 123 were aliens and 94 natives for the Czech Republic. For MRTs of these species see Tab. 3. Temporal changes of studied flora are expressed both by cummulative numbers of newly occurring species (Fig. 1) and rate of immigration expressed as number of new species divided by the length of the period (Fig. 2). The rate of immigration differs considerably between individual periods. Three immigration waves were distinguished, covering periods of (i) LBK - LNE, (ii) BR1 - LIA, and (iii) EM1 - EM2. The dynamism of immigration waves apparently swelled during time. Especially the species-richest Early Medieval Period, though short, represented an inception of the immediately continuing immigration flow since the High Middle Ages till the present. Potential causes of ascertained stagnation in ENE and RMP are discussed below.

186 140

120

100

number of species 40

0 LBK LNE ENE BR1 BR2 EIA LIA RMP EM1 EM2

kumul nativ kumul aliens

Fig. 1 Cummulative numbers of immigrated species plotted for aliens (kumul aliens) and natives (kumul nativ) against time of their immigration (for explanation of abbreviations see Tab. 1).

Fig. 2 Rate of immigration expressed per 100 years (the number of new immigrants divided by the length of period) according to time intervals of individual periods.

Conspicuously the waveforms of aliens and natives resemble mutually though the former species invaded gradually from remote areas, unlike natives spreading instantly from local vegetation. The proportion of alien species ranges in 42-55% of total species number in respective periods only. However, this resemblance vanishes when native species are divided to ecological specialists of natural habitats such as grasslands, and generalists which prevail in ruderal or segetal environment. The species of natural habitats rapidly escalate only since BR2

187 (Fig. 3), whereas proportions of ruderal/segetal native species are rather complementary to the representation of aliens in respective periods.

proportion (%) proportion 20

0 LBK LNE ENE BR1 BR2 EIA LIA RMP EM1 EM2 alien nativ rud nativ nat Fig. 3 Relative proportions of species groups, according to their presence in individual periods. The group of native species was divided into two groups: species of ruderalised habitats (nativ rud) and species of semi-natural habitats (nativ nat).

Fig. 4 PCA analysis showing distribution of taxa on first two axes.

188 Tab. 3 List of plant species. For each taxon, its occurrence in prehistoric period is shown. Species are divided into ecological groups.

Plant species LBK LNE ENE BR1 BR2 EIA LIA RMP EM1 EM2 status

LBK - LNE (Early Neolithic - Late Neolithic)

Weeds of contemporary fallow land and maize or vegetable plantations Chenopodium album x x x x x x x x x x Fallopia convolvulus x x x x x x x x x x Bromus secalinus x x x x x x x x x alien Galium aparine x x x x x x x x x x Galium spurium x x x x x x x x x x alien Chenopodium hybridum x x x x x x x x x x Persicaria lapathifolia agg. x x x x x x x x x x Solanum nigrum x x x x x x x x x x alien Chenopodium polyspermum x x x x x x x x Persicaria maculosa x x x x x x x x x x Atriplex patula x x x x x x x x x alien Bromus arvensis x x x x x x x x x alien Bromus sterilis x x x x x x x x x alien Echinochloa crus-galli x x x x x x x x x alien Lapsana communis x x x x x x x alien Setaria pumila x x x x x x x x x x alien Sinapis arvensis x x x x x x x alien Thlaspi arvense x x x x x x x alien Setaria viridis x x x x x x alien Euphorbia helioscopia x x x x x x x alien Digitaria ischaemum x x x x x x x alien Digitaria sanguinalis x x x x x x alien Lolium temulentum x x x x x x x x alien Stellaria media agg. x x x x x x x x Annual species of trampled and/or dunged bare soils Polygonum aviculare agg. x x x x x x x x x x Chenopodium murale x x x x x x x x x x alien Chenopodium urbicum x x x x x alien Poa annua x x x x x x Arenaria serpyllifolia agg. x x x x x x Capsella bursa-pastoris x x x x alien Perennial species of mezic ruderal grasslands Urtica dioica x x x x x x x x x Elymus repens x x x x x x x Medicago lupulina x x x x x x x Ranunculus repens x x x x x x x x Rumex acetosa x x x x x Silene vulgaris x x x x x x x Trifolium pratense x x x x x x x x x Vicia hirsuta/tetrasperma x x x x x x x x x Sambucus ebulus x x x x x x x x x alien Convolvulus arvensis x x x x x x x alien Securigera varia x x x x x Other species Atriplex sagittata x x x x x alien Papaver dubium/rhoeas x x x x x x alien Rumex acetosella x x x x x x x x x x Acinos arvensis x x x x x x Stipa pennata agg. x x x x x x

189 Plant species LBK LNE ENE BR1 BR2 EIA LIA RMP EM1 EM2 status

ENE – BR1 (Eneolithic, Early and Middle Bronze Age)

Annual cereal weeds Agrostemma githago x x x x x x x x alien Scleranthus annuus x x x x x x x x Veronica hederifolia agg. x x x x x x x Galeopsis tetrahit agg. x x x x x x x Anagallis arvensis agg. x x x x alien Stachys annua/arvensis x x x x x x x alien Viola cf. arvensis x x x x x x Buglossoides arvensis x x x x x x x alien Consolida regalis x x x x alien Polycnemum arvense x x x x x x x alien Adonis aestivalis x x x x x x alien Aethusa cynapium x x x Anthemis arvensis x x x x x x alien Apera spica-venti x alien Bupleurum rotundifolium x x x x x x x alien Fumaria officinalis x x x x x x x alien Fumaria vaillantii x x alien Kickxia elatine x x x alien Lolium remotum x alien Nigella arvensis x x x alien Silene noctiflora x x x x x alien Thymelaea passerina x x x x Veronica opaca/polita x x x alien Tall biennial and perennial herbs of dry and nitrogen-poor substrata Cirsium arvense x x x x x x alien Echium vulgare x x x x Silene latifolia x x x x x x x alien Galeopsis angustifolia/ladanum x x x x x x x Artemisia vulgaris x x x x Camelina microcarpa x x x x x x alien Daucus carota x x x x x x x Geranium columbinum x x x alien Lathyrus tuberosus x x alien Medicago falcata x x x x x x Melilotus albus/officinalis x x x x x x alien Reseda lutea x x x x alien Trifolium arvense x x x x x x Vicia cf. pannonica/sativa x x x x x x x alien Species of nutrient-rich trampled and grazed lawns Carex muricata agg. x x x x x x x Plantago lanceolata x x x x x x x x Rumex crispus/obtusifolius x x x x x x x x Trifolium campestre x x x x x Trifolium repens x x x x x x x x Galium mollugo x x x x x x x Glechoma hederacea x x x x x x Mentha arvensis x x x x x x x Potentilla anserina x x x x Prunella vulgaris x x x x x x Sonchus arvensis x x x alien Verbena officinalis x x x x x alien Other species Arctium sp. x x x x alien Centaurea jacea x x x x x x x Heracleum sphondylium x x x x

190 Plant species LBK LNE ENE BR1 BR2 EIA LIA RMP EM1 EM2 status

Raphanus raphanistrum x x x x x x alien Scirpus sylvaticus x x x x x Lamium amplexicaule/purpureum x x x x x x alien Lithospermum officinale x x x x x Atriplex oblongifolia x x x x alien Galium verum agg. x x x x x Glaucium corniculatum x x x x alien Hyoscyamus niger x x x x x x alien Leucanthemum vulgare aggr. x x x x x Malva neglecta x x x x x x alien Mercurialis annua x x x alien Senecio vulgaris x x x alien Trisetum flavescens x x x x x x

BR2, EIA, LIA, RMP – Late and Final Bronze Age, Early and Late Iron Age, Roman and Migration period

Ruderal species of sunny, unevenly disturbed substrata rich in bases and organic nutrients Bromus tectorum x x alien Bryonia alba x alien Erodium cicutarium x alien Chenopodium bonus- henricus x alien Malva sylvestris x x x x alien Anthemis cotula x x x alien Carduus acanthoides x x x alien Geranium pusillum x x alien Lepidium campestre x x x alien Malva pusilla x x x alien Onopordum acanthium x x x alien Stachys germanica x x x alien Urtica urens x x x alien Species of meadows, pastures and dry grasslands Stellaria graminea x x x x Ajuga genevensis/reptans x x x x x Vicia cracca/sepium x x x x x x Carex hirta x x x x x x Carex leporina x x x x x Phleum pratense x x x x x Barbarea vulgaris x x x Centaurea scabiosa x x x x x Clinopodium vulgare x x x x Hypericum perforatum x x x x x Lychnis flos-cuculi x x x x x Plantago media x x x x Thalictrum minus x x x x Pimpinella saxifraga x x x Carex pallescens x x x Ranunculus acris x x x x x Silene nutans x x x Knautia arvensis x x x Linum catharticum x x x Melampyrum arvense x x x x alien Taraxacum sp. x x x Other species Chenopodium ficifolium x x x x x x Chenopodium glaucum/rubrum x x x x x x Valerianella dentata x x x x x alien

191 Plant species LBK LNE ENE BR1 BR2 EIA LIA RMP EM1 EM2 status

Avena fatua x x x x x alien Potentilla sp. x x x x x Descurainia sophia x x x alien Galium tricornutum x x x alien Lepidium ruderale x x x x alien Odontites vernus x x x Papaver argemone x x x alien Stachys palustris x x x Ajuga chamaepitys x x alien Asperula arvensis x x x alien Hibiscus trionum x alien Chelidonium majus x x x alien Chenopodium foliosum x alien Matricaria maritima x alien Myosotis arvensis x x x x alien Aegopodium podagraria x x Aphanes arvensis x x x Neslia paniculata x x x x alien Valerianella rimosa x alien

EM1, EM2 – Early Middle Ages 1-3, Early Middle Ages 4

Nitrophilous ruderal species of human-made substrata Lamium album x x alien Ballota nigra x x alien Setaria verticillata/viridis x x alien Conium maculatum x x alien Leonurus cardiaca x x alien Xanthium strumarium x x alien Sonchus asper x x alien Sonchus oleraceus x x alien Amaranthus blitum/graecizans x x alien Atriplex prostrata x x Euphorbia peplus x x alien Hordeum murinum x alien Nepeta cataria x x alien Rumex conglomeratus x Portulaca oleracea x alien Lactuca serriola x alien Anthriscus caucalis x alien Chenopodium vulvaria x alien Sisymbrium officinale x alien Pastoral species avoided by animal grazing Cerinthe minor x x Agrimonia eupatoria x x Linaria vulgaris x x alien Cichorium intybus x x alien Berteroa incana x x alien Anchusa officinalis x x alien Cirsium vulgare x x Euphorbia cyparissias x x Marrubium vulgare x alien Species of wet forests and alluvial meadows Fallopia dumetorum x x Pastinaca sativa x x Ranunculus flammula x x Silene dioica x x Thalictrum flavum x x Valeriana officinalis x x

192 Plant species LBK LNE ENE BR1 BR2 EIA LIA RMP EM1 EM2 status

Rumex sanguineus x x Stellaria palustris x x Filipendula ulmaria x Other species Geranium dissectum x x alien Centaurea cyanus x x alien Reseda luteola x x alien Caucalis platycarpos x x alien Spergula arvensis x alien Ranunculus arvensis x alien Silene dichotoma x alien Anthemis austriaca x alien Crepis capillaris x alien Geranium molle x alien MIcrorrhinum minus x alien Silene gallica x alien Vaccaria hispanica x alien

Temporal changes of the species composition

The floristic resemblance of individual periods based on their species composition is expressed on Fig. 4. The first axis can be understood as a gradient between R and CRS strategists, which is linked to the periodicity of disturbances. Its left side corresponds to prevailing one-shot intensive impacts resulting in bare soils whereas the right side apparently comprises a long- term methodical management with regular disturbances. The second axis, on the other hand, represents the moisture gradient. Thus, habitat preferences of represented species imply, according to the individual quadrants, following types of environment: (i) moderately humid and, at the same time, disturbed open soils; (ii) dry and disturbed open soils; (iii) arable fields of cereals; and (iv) mowed meadows or pastured grasslands. Furthermore, studied periods can be classificated into four chronologic phases according to their species composition.

Temporal changes of the vegetation based on the first emergences of ecological groups

The relation between MRTs and habitat demands of individual species is included in Tab. 3. The results enabled a more detailed view into vegetation ecology of the ancient cultural landscape. We distinguished following nine species-groups, which newly emerged during the four developmental phases. Note, that the delimited phases (LBK-LNE, ENE-BR1, BR2-RMP, EM1-EM2) differ from those in Fig. 4, since they are defined by just immigrating species, whereas the latter are based on total floristic composition (LBK, LNE-ENE, BR1-RMP, EM1- EM2).

Phase 1 (LBK-LNE)

Weeds of contemporary fallow land and maize or vegetable plantations Species of nutrient-rich, frequently disturbed soils are common. The group comprises annual dicots (Chenopodium album, Fallopia convolvulus, Persicaria lapathifolia), winter annual grasses (Bromus arvensis, B. secalinus, B. sterilis) and tardily germinating annual grasses with autumnal optimum, mostly having C4 metabolism (Digitaria ischaemum, D. sanguinalis, E.

193 crus-gali, Setaria pumila, S. viridis). At present, most these species occur in sandy-loamy grounds of ruderal habitats, fields and gardens. Furthermore, aliens such as Atriplex patula, Euphorbia helioscopia, Sinapis arvensis, Solanum nigrum, Thlaspi arvense and natives such as Chenopodium polyspermum, Stellaria media indicate repeatedly disturbed, nutrient-rich soils occuring at present in hoed root crops or vegetable gardens. This species composition clearly indicates weed vegetation of phytosociological units Spergulo arvensis-Erodion cicutariae and Veronico-Euphorbion. These units often form mixed stands as autumnal and vernal phenological vicariants, respectively.

Annual species of trampled and/or dunged bare soils The group includes ruderal species of bare soils which are trampled (Capsella bursa-pastoris, Polygonum aviculare, Poa annua,) and/or supplied by ammoniacal nitrogene (Chenopodium hybridum, C. urbicum, C. murale). In modern times, this species combination typically occurs in open grounds along buildings, walls, and various human defecation/urination sites, sites for livestock breeding as well as near dung holes.

Perennial species of mezic ruderal grasslands This group is formed by generalistic species which are widespread in current landscape and easily colonize periurban, industrial or minning areas, road embankments or balks separating fields. These successional stages mostly last for several tens of years and lack regular management excluding occassional events of trampling, cutting or deposition of waste. They are often dominated by rhizomate geophytes, e.g. Convolvulus arvensis, Elymus repens, Sambucus ebulus, Urtica dioica.

Phase 2 (ENE-BR1)

Annual cereal weeds Thermophilous annual species of cereal fields (Adonis aestivalis, Agrostemma githago, Bupleurum rotundifolium, Camelina microcarpa, Consolida regalis, Nigella arvensis) and stubble fields (Stachys annua/arvensis, Anagallis arvensis, Buglossoides arvensis, Polycnemum arvense, Silene noctiflora, Kickxia elatine) clearly correspond to the unit Caucalidion which represents field weed vegetation of fertile soils rich in mineral nutrients such as chernozem. Similarly, a group of less nutrient-demanding species of base-poor soils (aliens such as Raphanus raphanistrum, Apera. spica-venti, Lolium remotum, Anthemis arvensis; natives such as Scleranthus annus, Galeopsis. tetrahit agg., Mentha arvensis) may indicate the unit Scleranthion i.e. weed vegetation of less fertile corn fields.

Tall biennial and perennial herbs of dry and nitrogen-poor substrata This species-rich group includes alien rhizomatous geophytes (e.g. Cirsium arvense, Sonchus arvensis, Lathyrus tuberosus) and short-lived herbs which are both of alien origin (e.g. Silene latifolia, Melilotus albus/officinalis, Atriplex oblongifolia, Vicia sp., Reseda lutea), and natives (e.g. Daucus carota, Artemisia vulgaris, Echium vulgare, Galeopsis angustifolia/ ladanum). These species indicate dry and sunny sites with bare loamy to stony soils which are rich in mineral nutrients but poor in phosphorus and nitrogene. In modern landscape, this species composition corresponds to unit Dauco-Melilotion, the vegetation type resulting from heavy

194 disturbances followed by several years of succession, e.g. in newly abandoned fields, road margins, eroded slopes or stone pits.

Species of nutrient-rich trampled and grazed lawns Species such as Trifolium campestre, T. repens, Carex muricata, Plantago lanceolata, Potentilla anserina, Glechoma hederacea, Prunella vulgaris, Verbena officinalis, and Rumex crispus/obtusifolius represent short ruderalised grasslands. Most of the species are native. In modern times, such vegetation occurs in villages (Potentilion anserinae, pastures at compacted and nitrified soils, mostly maintained by poultry or goats) or in their vincinity (Alchemillo- Ranunculion repentis, eutrophic short lawns).

Phase 3 (Br2, EIA, LIA, RMP)

Ruderal species of sunny, unevenly disturbed substrata rich in bases and organic nutrients Today’s co-occurrence of these species results from early succession on bare or repeatedly disturbed soils around sites of animal breeding. Tall xerophilous and thermophilous biennial herbs (Onopordon acanthium, Carduus acanthoides, Stachys germanica) combined with short annuals (Bromus tectorum, Lepidium ruderale, Erodium cicutarium) may indicate the unit Onopordion. Similar, but moderately wet and highly nitrified habitats may be indicated by species of today’s units Malvion neglectae (low annuals such as Urtica urens, Malva pusilla, Anthemis cotula, Chenopodium. glaucum/rubrum), and Arction (tall perennials Malva sylvestris, Bryonia alba, C. bonus-henricus). In rural settlements, these three units often form a tessellate pattern which is spatially differentiated according to quality and intensity of human or animal impact.

Species of meadows, pastures and dry grasslands All these species are native in central Europe. Combination of species linked to mezic or wet habitats (such as Carex leporina, C. pallescens, C. hirta, L. flos-cuculi, Phleum pratense, Stachys palustris) may indicate spasmodic rotation of disturbance events (e.g. poaching by cattle) and periods of abandonment. Species such as Plantago media, Pimpinella saxifraga, Centaurea scabiosa, Clinopodium vulgare, Silene nutans, Hypericum perforatum, Thalictrum minus indicate dry grasslands, shrubby fringes or, possibly, pastural forests.

Phase 4 (EM1, EM2)

Nitrophilous ruderal species of human-made substrata This group includes mesophilous and nitrophilous weeds demanding or toleranting high content of nitrogene and other nutrients. Tall and robust weeds (Ballota nigra, Conium maculatum, Leonurus cardiaca, Nepeta cataria) correspond to nutrient-rich substrata, undisturbed for several years. Short annuals such as Anthriscus caucalis, Chenopodium vulvaria, Xanthium strumarium prefer intensively disturbed sites rich in ammonia nitrogene. Tall annuals (Lactuca serriola, Sisymbrium officinale, Sonchus asper, S. oleraceus) are, in modern times, common at rubbish dumps or refuse heaps. These three vegetation types can be associated to the units of Arction, Malvion neglectae, and Atriplicion, respectively.

195 Pastoral species avoided by animal grazing Indigestible, aromatic or poisonous pastural weeds emerged in this period. Aliens such as Anchusa officinalis, Berteroa incana, Cichorium intybus, Linaria vulgaris and natives such as Agrimonia eupatoria, Cerinthe minor, Cirsium vulgare, Euphorbia cyparissias indicate pastural degradation of mezic or dry grasslands.

Species of wet forests and alluvial meadows This species group indicate wet nad nutrient-rich habitats of fluvial plains or spring beds. Tall herbs such as Thalictrum flavum, Valeriana officinalis and Filipendula ulmaria indicate meadows whereas Fallopia dumetorum, Silene dioica and Rumex sanguineus may reveal forest edges or clearings.

Discussion

The nature of our data and consequences for their interpretation

The understanding of the invasion process needs detailed taxonomic identification of invaders, reliable assessment of their residence times, and analysis of their individual invasion histories in context of natural and socioeconomic changes of a landscape. This knowledge is already advanced, as far as modern invaders are concerned (Pyšek et al. 2003), unlike the archaeophytes. Study of their past invasion requires assessment of fossil record which is always less reliable than data from the present or modern times. Therefore we are aware of many possible imperfections owing both identification of studied plant remains and interference of various natural and taphonomical processes (see above). However, our work represents just the first attempt to synthetize extensive macroremain data from the whole Czech Republic. Similar works from other countries are not available for close analogy owing to the regional differences (e.g. Preston et al., Brun 2009) or partly different approach (for Germany: Willerding 1986, Poschlod 2015; for Poland: Trzcinska-Tacik–Wasylikowa 1982, Litynska-Zajac 2005) .

Further ahead, special evaluations of statistic, archaeological or taphonomical aspects of our material would be necessary in separate papers. In this work we only wish to open this theme and show a first look to the problems. Therefore we targeted only to rough and general comments of flagrant findings which may adumbrate orientation of further detailed investigation.

Course from populated spots to connected landscape

Our basic finding is the identification of progressive, though fluctuating increase of species diversity since the Neolithic to the beginning of the Medieval Period. We try to explain this trend as a result of the increase of species diversity, partly in agricultural enclaves, partly in palaeorecord.

The species diversity grew with gradual expansion of agricultural enclaves and particularly their interconnecting, which facilitated redisposition of materials, human migration and plant dispersal. Further contributing factor was gradational growth of cultural diversity resulting e.g. in different settlement forms or variety of co-existing management styles in which new habitats emerge, new niches were opened and finally new vegetation types developed.

196 However, the species diversity likewise increases from the taphonomic reasons. The Neolithic assemblages show dominance of annual weeds, whereas the grassland flora affiliates only later and became gradually dominant (Fig. 3). It is nevertheless clear that grasslands surrounded the settlements during all periods, and especially in the Neolithic, when the populated enclaves were situated instantly within the nature, the prevalence of grassland or also forest species over weeds could be expected. Similarly, the species of wetlands appear only individually until the medieval period, though some wet habitats were always within easy reach. This contradiction can be interpreted by gradually increasing role of specialized human activities, which enabled that more and more species entered into archeologized material.

Last not least, the diversity is ruled by total data amount in individual periods. The number of investigated sites varied between 10 in LBK and in 46 BR2 or 36 in EM2, but much more significant we consider huge differences in number of recorded seeds. The seed sum is not reaching 6,000 until BR2, when it rises to ca 21,000, and in EM1 and EM2 it exceeds 70,000 (Tab.1).

Early segetal vegetation: cultivated natives or alien weeds?

As a conspicuous phenomenon we consider the starting phase of agricultural weed flora occurred ahead of the wave of specialized cereal weeds, taking place since the Eneolithic. The difference between the two species groups was probably caused by soil quality and local management. The weeds occurring in neolithic fields are today common, partly in hoed and manured plantations, and partly in bare soils such as loess or loam in which nitrogen is low but very well accessible.

Most grasses presumably growing as segetal weeds during the Neolithic posse large edible seeds, and some of them were documented to be utilized as substitutes of cereals in some countries in the past, viz. Echinochloa crus-gali, Bromus secalinus, and Setaria pumila. The same is known also for the native dicot species such as Chenopodium album, Fallopia convolvulus and Persicaria lapathifolia (Behre 2008). Especially interesting are finds of Fallopia convolvulus in Mesolithic settlements in Sweden (Regnell 2012), Scottland (Bishop et al. 2014), Russia (Dolukhanov 2016) and Belgium (Crombé et al. 2015), since this particular species is considered to be alien in the Czech Republic. These finds may imply some pre- neolithic experiences of humans with these species.

Development of ruderal vegetation and grasslands

Since this vegetation is considerably spatially diversified, we may expect that many various vegetation types coexisted within the same settlement, and that they, besides, changed during the time in dependence to the character of human impact. Especially the succesion of ruderal vegetation is not easily interpretable without special comparison to archeological knowledge. Sequence of Dauco-Melilotion - Onopordion - Arction communities (see results) is a possible exclusion. This chain can be interpreted by a gradual rise of nutrients in the settlement soils since the Eneolithic to the Medieval period, which may be linked e.g. to the growing intensity of human impact or to the cumulative effect due to duration of these habitats.

197 The grassland species are present in our material since the LBK phase, since they either remained around the settlements from the original vegetation, or they newly established at disturbed sites. The first apparent display of grasslands shows typical short lawns. According to their species composition, they resemble to the sites of current rural settlements influenced by systematic pasture and manure of domestic poultry. Sudden rise of more specialised grassland species in BR2 (Fig. 3) is apparently a result of then emerged hayfields, accompanied by typical meadow species (Tab. 3). Their arrival corresponds well with the first finds of scythes in late Bronze excavations (Jiráň 2008).

Development of species composition

Most of the herbal archaeophytes, expected to be introduced unintentionally to the Czech Republic, were found in the studied material. From 290 species of archaeophytes listed in current Czech flora (Danihelka et al. 2012), 240 species obviously have never been planted intentionally. From this number, 217 archaeophytes were included to this study, representing ca. 90% of them. The remaining ca. 20 species were not detected archaeobotanically, or they immigrated later, i.e. during the High Middle Ages.

All the species found are listed in current Czech flora. However, in many species it still remains questionable whether they survived continually after their introduction or whether their historical occurrence results from independent immigration events. This is the case of e.g. Chenopodium foliosum which was found rarely in EIA, and its present status is ‘ocassionally cultivated and casually escaping plant’ (Pyšek et al. 2012).

The species composition of concerned archaeological sites differs considerably from the present-day flora of human-made habitats, especially by their low proportion of native species. The suppression of natives can be explained by following factors: (i) Many natives, such as tiny vernal annuals or plants of extreme habitats remained ever out of human treatment; (ii) cereal weeds are better adapted to persist in grain material (De Wet–Harlan 1975) than easily separated natives; (iii) in many natives of natural habitats, seed production is low, especially under human impact. Owing this distortion of data, many putative vegetation patterns and processes do not prove, for example, no clear responses to Holocene climatic changes were registered.

The relationship between numbers of native and alien species is positive among data from individual periods (see Fig. 1). This feature, however, is not very surprising. This relation is strongly negative in communities with low and fixed number of species, but it changes owing to scale-dependence (Herben et al. 2004). We consider the similar numbers of aliens and native species being a result of heterogeneity of our data. Individual periods subsume various numbers of assemblages coming from different landscape situations, and each assemblage represents a different habitat mosaic. Different taphonomical history of fossil record, e.g. carbonized weeds in cereal stockpiles vs. watterlogged waste material may be further source of data heterogeneity.

Acknowledgement This study was supported by the project No. 13–11193S of the Czech Science Foundation (GA ČR), by the project No. 13–08169S of the Czech Science Foundation (GA ČR), by the project No. LO1305 of the Ministry of Education, Youth and Sports of the Czech Republic, and by the PAPAVER project No. cz.I.07 /2.3.00/20.0289 of the Ministry of

198 Education, Youth and Sports of the Czech Republic co–financed by the European Social Fund and the state budget of the Czech Republic. The authors would like to thank to all archaeologists who support archaeobotanical investigation in their research sites. We also would like to thank Věra Čulíková for her kindly help with searching of literature for this work and also for her consultations for determination of problematic finds.

References

Behre KE (2008) Collected seeds and fruits from herbs as prehistoric food. Vegetat Hist Archaeobot 17(1):65–73

Behre KE, Jacomet S (1991) The ecological interpretation of archaeobotanical data. In: van ZeistW,Wasylikowa K, Behre KE (eds.) Progress in Old World Palaeoethnobotany. 81–108, Balkema, Rotterdam-Brookfield

Berkovec T, Kočár P, Kočárová R (2005) Archeobotanický výzkum na lokalitě Kroměříž, Újezd u svatého Františka. In: Bém M, Peška J (eds.) Ročenka 2004. 94–125, Archeologické Centrum Olomouc, Olomouc

Bieniek A (2002) Archaeobotanical analysis of some early Neolithic settlements in the Kujawy region, central Poland, with potential plant gathering activities emphasised. Vegetation history and archaeobotany 11(1–2):33–40

Bishop RR, Church MJ, Rowley-Conwy PA (2014) Seeds, fruits and nuts in the Scottish Mesolithic. Proceedings of the Society of Antiquaries of Scotland 143

Bosi G, Bandini Mazzanti M, Florenzano A, Massamba N’siala I, Pederzoli A, Rinaldi R, Torri P, Mercuri AM (2011) Seeds/fruits, pollen and parasite remains as evidence of site function: Piazza Garibaldi e Parma (N Italy) in Roman and Mediaeval times. Journal of Archaeological Science 38:1,621–1,633

Braun-Blanquet J (1964) Pflanzensoziologie, Grundzüge der Vegetationskunde. Vienna

Brun C (2009) Biodiversity changes in highly anthropogenic environments (cultivated and ruderal) since the Neolithic in eastern France. The Holocene 19(6):861–871

Chvojka O, Beneš J, Fröhlich J, John J, Michálek J, Šálková T, Bernardová A, Houfková P, Křivánek R, Majer A, Novák J, Nováková K, Talarovičová E (2011) Osídlení doby bronzové v povodí říčky Smutné v jižních Čechách. Archeologické výzkumy v jižních Čechách, Supplementum 8, České Budějovice

Chvojka O, Menšík P, Prokop V, Novák J, Šálková T, Fröhlich J, Michálek J (2014) Sídliště z mladší doby bronzové na Zhoři na Táborsku. Archeologické výzkumy v jižních Čechách 27:69–85

Chytráček M, Danielisová A, Pokorný P, Šmejda L, Kočár P, Kočárová R, Kyselý R, Kyncl T, Sádlo J, Zavřel J (2012) Vzestupy a pády regionálního mocenského centra. Přehled současného

199 stavu poznání pravěkého opevněného areálu na Vladaři v západních Čechách. Památky archeologické 103:273–338

Chytráček M, Šmejda L, Danielisová A, Pokorný P, Křivánek R, Kočár P, Klsák J (2010) Blockbalkenkonstruktionen des 5. Jahrhunderts v. Chr. im feuchtbodenmilieu der Vorburg des Burgwalls Vladař in Westböhmen. In: Husty L (ed.) Fines Transire 19: 183–192

Crombé P, Langohr R, Louwagie G (2015) Mesolithic hearth-pits: fact or fantasy? A reassessment based on the evidence from the sites of Doel and Verrebroek (Belgium). Journal of Archaeological Science 61:158–171

Čulíková V (1986) Rekonstrukce synantropní vegetace středověkého Mostu na základě makrozbytků z antropogenních uloženin. Disertační práce

Čulíková V (1998) Rostlinné makrozbytky z raně středověkých sedimentů na III. nádvoří Pražského hradu. Archaeologica Pragensia 14:329–341

Čulíková V (1998a) Výsledky analýzy rostlinných makrozbytků z lokality Praha 1 - Malá Strana, Tržiště čp 259/III (Hartigovský palác). Archaeologica Pragensia 14:291–316

Čulíková V (2001) Rostlinné makrozbytky z pěti středověkých lokalit při obvodu centrální části Pražského hradu. Medievalia archaeologica 3:303–327

Čulíková V (2001a) Rostlinné makrozbytky z lokality Praha1- Malá Strana, Malostranské nám. čp. 158/III. (Lichtenštejnský palác). Medievalia archaeologica 3:137–166

Čulíková V (2005) Rostlinné makrozbytky z raně středověké lokality Mostecká – Josefská ul. (dřevěná cesta), Praha 1 – Malá Strana. Archeologica Pragensia 17:137–169

Čulíková V (2008) Rostlinné makrozbytky z pravěkých a raně středověkých antropogenních sedimentů v Lovosicích. Archeologické rozhledy 60:61–74

Čulíková V (2009) Rostlinné makrozbytky. In: Zápotocký M, Zápotocká M (eds.) Kutná Hora – Denemark. Hradiště řivnáčské kultury (ca 3000–2800 př.Kr.). Památky archeologické, Supplementum 18:255–264

Čulíková V (2010) Středověká údolní niva Vltavy v Praze na Malé Straně (Valdštejnská čp. 154/III, Kolovratský palác). Archeologické rozhledy 62:72–116

Danielisová A, Langová M, Kočár P, Kyselý R, Stránská P, Sůvová Z, Světlík I (2013) Mohyla únětické kultury z Brandýsa nad Labem jako doklad ojedinělých pohřebních praktik starší doby bronzové. Archeologické rozhledy 65:56–88

Danihelka J, Chrtek Jr J, Kaplan Z (2012) Checklist of vascular plants of the Czech Republic. Preslia 84(3):647–811

200 De Wet, J. MJ, Harlan JR (1975) Weeds and domesticates: evolution in the man-made habitat. Economic Botany 29(2):99–108

Dolukhanov PM (2016) 15 Plants and subsistence of huntergatherers in the prehistoric East European Plain (Upper Palaeolithic, Mesolithic and SubNeolithic). Hunter-Gatherer Archaeobotany: Perspectives from the Northern Temperate Zone.

Dreslerová D, Kočár P (2013) Trends in cereal cultivation in the Czech Republic from the Neolithic to the Migration period (5500 B. C. – A.D. 580). Veget Hist Archaeobot 22:257–268

Dreslerová D, Kočár P, Chuman T, Pokorná A (in press) Cultivation with deliberation: cereals and their growing conditions in prehistory. Veget Hist Archaeobot

Dreslerová D, Sádlo J (2000) Les jako součást pravěké kulturní krajiny [The Forest as a Component of the Prehistoric Cultural Landscape]. Archeologické rozhledy 52:330–346

Dreslerová D, Waldhauser J, Abraham V, Kočár P, Křivánek R, Meduna P, Sádlo J (2013) Bezdězsko – Dokesko v pravěku a laténské sídliště v Oknech. Archeologické rozhledy 65:535– 573

Ernée M, Dobeš M, Hlaváč J, Kočár P, Kyselý R, Šída P (2007) Zahloubená chata ze středního eneolitu v Praze 9 – Miškovicích. Výsledky archeologických a přírodovědných analýz. Památky archeologické 97:31–108

Fietz A (1941) Bronzezeitliche Pflanzenreste aus der Stierfelshöhle bei Brünn. –Verhandlungen des Naturforschenden Vereines in Brünn 72:71–75

Fröhlich J (2004) Časnělaténská polozemnice v Oseku u Milevska. Archeologické výzkumy v jižních Čechách 17:185–196

Greig J (1981) The investigation of a medieval barrel-latrine fromWorcester. Journal of Archaeological Science 8:265–282

Heimdahl J (2005) Urbanised Nature in the Past Site Formation and Environmental Development in Two Swedish Towns, AD 1200e1800. PhD thesis, Stockholm University, Department of Physical Geography and Quaternary Geology

Herben T, Mandák B, Bímová K, Münzbergová Z (2004) Invasibility and species richness of a community: a neutral model and a survey of published data. Ecology 85(12):3,223–3,233

Hlásek D, Houfková P, Kovačiková L, Majer A, Novák J, Pavelka J, Bešta T, Šálková T (2014) The Use of Environmental Methods for Studying the Fortifcation, Economic System and Natural Environment of a Hillfort from the Beginning of the Middle Bronze Age at Vrcovice, Czech Republic. Interdisciplinaria Archaeologica Natural Sciences in Archaeology 5(1):31–47

201 Jiráň L (2008) Archeologie pravěkých Čech 5. Doba bronzová. Archeologický ústav AVČR, Praha

Jiráň L, Venclová N (2007–2008) Archeologie pravěkých Čech, díl 1-8. Archeologický ústav AV ČR, Praha, v.v.i.

Jiřík J, Pták M, Šálková T, Tisucká M (2012) Pohřbívání v Písku na Bakalářích. Acta archaeologica Opavensia 4:113–118

John J, Eigner J, Fröhlich J, Šálková T (2012) Výšinné sídliště u Velké a další nové poznatky o eneolitických výšinných lokalitách na střední Vltavě. Archeologické výzkumy v jižních Čechách 25:63–82

Kalábek M, Kalábková P, Kočár P, Kočárová R (2010) Největší lengyelská stavba z Hulína– Pravčic (struktura 45). In: Bém M, Peška J (eds.) Ročenka 2009:42–58, Archeologické Centrum Olomouc, Olomouc

Kalábek M, Kočár P (2007) Laténský objekt s nálezy obilí z lokality Olomouc–Řepčín 1. Pramen k poznání zemědělství v mladší době železné. In: Bém M, Peška J (eds.) Ročenka 2006:94–125, AC Olomouc, Olomouc

Kalis AJ, Merkt J, Wunderlich J (2003) Environmental changes during the Holocene climatic optimum in central Europe-human impact and natural causes. Quaternary Science Reviews 22(1):33–79

Klečka A, Skutil J (1937) Moravské příspěvky k výzkumu pravěkých obilnin. Věstník československého zemědělského muzea 1:19–25

Kočár P, Kočárová R (2007) Rostlinné zbytky z mladobronzových lokalit na katastru obce Tuchoměřice. In: Salaš M, Šabatová K (eds) Doba popelnicových polí a doba halštatská: Příspěvky z IX. konference, Bučovice 3.–6.10.2006. 305–313, Masarykova univerzita, Brno

Kočár P, Kočárová R (2012) A o rostlinách. In: Šumberová R (ed.) Cesta napříč časem a krajinou. 71–80, Archeologický ústav AV ČR, Praha, v.v.i.

Kočár P, Mihályiová J (2011) Kněževes – archeobotanická analýza na sídlišti z mladší doby bronzové. In: Smejtek L (ed.) Osídlení z doby bronzové v Kněževsi u Prahy. 464–484, Ústav archeologické a památkové péče středních Čech, Praha

Kreuz A, Schäfer E (2002) A new archaeobotanical database program. Veget Hist Archaeobot 11:177–179

Kühn F (1960) Nálezy obilnin z pravěkých výzkumů. Archeologické rozhledy 12:701–708

202 Kühn F (1972) Obilí z halštatské doby z Býčí skály u Adamova. Vědecké práce Zemědělského muzea 12:9–22

Kühn F (1980) Botanický rozbor obilí z halštatské doby z Rajhradu (okr. Brno–venkov). Přehled výzkumů 1977:43–46

Kühn F (1981a) Botanický rozbor obilí z eneolitu z Bořitova. Přehled výzkumů 1979:20

Kühn F (1981b) Rozbor nálezů polních plodin. Přehled výzkumů 1979:75–79

Kühn F (1981c) Crops and weeds in Šlapanice near Brno from Bronze Age to now. Zeitschr. Archaeol. 15:191–198

La Sorte FA, Pyšek P (2009) Extra‐ regional residence time as a correlate of plant invasiveness: European archaeophytes in North America. Ecology 90(9):2,589–2,597

Lityńska-Zając M (2005) Chwasty w uprawach roślinnych w pradziejach i wczesnym średniowieczu. Instytut Archeologii i Etnologii Polskiej Akademii Nauk

Nekvasil J, Opravil E (1994) Dvě starolužické jámy z Loštic. Severní Morava 68:3–11

Neustupný E (2008) Archeologie pravěkých Čech 4. Eneolit. Archeologický ústav AVČR, Praha

Opravil E (1972) Rostliny z velkomoravského hradiště v Mikulčicích. Archeologický ústav ČSAV v Brně, Academia, Praha

Opravil E (1976) Archeobotanické nálezy z městského jádra Uherského Brodu. Studie Archeologického ústavu ČSAV 1974 3(4)

Opravil E (1979) Rostlinné zbytky z Mohelnice 1, 2. Časopis Slezského muzea A 28:1–13, 97– 109

Opravil E (1980) Rostlinné nálezy ze středověku Starého Města (okr. Uherské Hradiště). Přehled výzkumů 1977: Archeologický ústav ČSAV, Brno 103–105

Opravil E (1985) Rostliny z mladší doby hradištní z Olomouce (okr. Olomouc). Přehled výzkumů 1983: Archeologický ústav ČSAV, Brno 51–54

Opravil E (1986) Rostlinné makrozbytky z historického jádra Prahy. Archaeologica Pragensia 7:237–271

Opravil E (1989) Zuhelnatělé obilí z doby římské z Drnholce (okr. Břeclav). Přehled výzkumů 1986:111–112

Opravil E (1992) Rostlinné zbytky ze středověkého hradu v Hradci nad Moravicí z let 1985, 1986 a 1988. Opravil, E. (1992). Rostlinné zbytky ze středověkého hradu v Hradci nad Moravicí. Časopis Slezského muzea, A41:97–104

203 Opravil E (1994) Synantropní vegetace ze středověku a z počátku novověku města Olomouce. Zprávy České botanické společnosti 11:15–36

Opravil E (1998) Plant macro–residues. In: Venclová N (ed.) Mšecké Žehrovice in Bohemia. Chronothëque 2:295–304

Opravil E (1998) Zusammenfassende Ubersicht der Ergebnisse von Analysen der Makroreste pflanzlicher Herkunft aus Mikulčice. Studien zum Burgwall von Mikulčice 3:327–353

Opravil E (2002) Rekonstrukce vegetace podle makrozbytků v oblasti pod Pavlovskými vrchy. In: Stuchlík S (ed.) Oblast vodního díla Nové Mlýny od pravěku do středověku. Spisy Archeologického ústavu AV ČR Brno 20:459–464

Pavlů I, Zápotocká M (2007) Archeologie pravěkých Čech 3. Neolit. Archeologický ústav AVČR, Praha

Pokorná A, Dreslerová D, Křivánková D (2011) Archaeobotanical Database of the Czech Republic, an Interim Report. Interdisciplinaria Archaeologica. Natural Sciences in Archaeology 1:49–53

Pokorný P, Boenke N, Chytráček M, Nováková K, Sádlo J, Veselý J, Kuneš P, Jankovská V (2006) Insight into the environment of a pre–Roman Iron Age hillfort at Vladař, Czech Republic, using a multi–proxy approach. Veget Hist Archaeobot 15:419–433

Pokorný P, Chytrý M, Juřičková L, Sádlo J, Novák J, Ložek V (2015) Mid-Holocene bottleneck for central European dry grasslands: Did steppe survive the forest optimum in northern Bohemia, Czech Republic? The Holocene 25:716–726

Poschlod P (2015) Geschichte der Kulturlandschaft. Verlag Eugen Ulmer, Stuttgart (Hohenheim)

Preston CD, Pearman DA, Hall AR (2004) Archaeophytes in Britain. Botanical Journal of the Linnean Society 145:257–94

Pyšek P, Jarošík V (2005) Residence time determines the distribution of alien plants. In: Inderjit S (ed.) Invasive plants: ecological and agricultural aspects. 77–96, Birkhäuser Basel.

Pyšek P, Sádlo J, Mandák B, Jarošík V (2003) Czech alien flora and the historical pattern of its formation: what came first to Central Europe? Oecologia 135(1):122–130

Regnell M (2012) Plant subsistence and environment at the Mesolithic site Tågerup, southern Sweden: new insights on the “Nut Age”. Vegetat Hist Archaeobot 21(1):1–16

Richardson DM, Pyšek P (2006) Plant invasions: merging the concepts of species invasiveness and community invasibility. Progress in Physical Geography 30(3):409–431

204 Rulf J, Tempír Z (2002) Bylany – ekodata: Nálezy zuhelnatělých obilek a semen. In: Pavlů I (ed.) Bylany Varia 2: 117–123, Institute of Archaeology in Prague, v.v.i., Praha

Šabatová K, Kazdová E, Kočár P, Kuča M, Nývltová–Fišáková M, Prokeš L, Sázelová S, Trampota F (2012) Ein Befund mit neolithischen Plastiken aus Těšetice–Kyjovice, Mähren. Fundverteilungsanalyse. – In: Link T, Schimmelpfennig D (eds.) Taphonomie. Beiträge zur Sitzung der AG Neolithikum im Rahmen der 80: 201–219, Verbandstagung des West– und Süddeutschen Verbands für Altertumsforschung e. V Fokus Jungsteinzeit – Berichte der AG Neolithikum 3 (Kerpen–Loogh in Vorbereitung), Nürnberg

Salaš M, Jarošová I, Kočár P, Nývltová Fišáková M, Roblíčková M (2012) Potravní zdroje obyvatelstva mladší doby bronzové na Cezavách u Blučiny. Analýzy bioarcheologických pramenů. Archeologické rozhledy 64:391–442

Šmejda L, Chytráček M, Pokorný P, Kočár P (2010) Nové poznatky z výzkumu hradiště Vladař u Žlutic, okr. Karlovy Vary. In: Krištuf P, Vařeka P (eds.) Opomíjená archeologie 2007– 2008:46–53, University of West Bohemia, Plzeň

Šmejda L, Kočár P (2007) Botanické makrozbytky z knovízského sídliště v Praze–Hostivaři: vektorová syntéza dat. In: Krištuf P, Šmejda L, Vařeka P (eds.) Opomíjená archeologie 2005– 2006:192–208, Elce Book Publishing, Plzeň

Tempír Z (1961) Archeologické nálezy obilnin na území Československa. Vědecké práce ústředí pro studium vývoje v zemědělství ČSAZV– Zemědělské muzeum v Praze 1:159–200

Tempír Z (1966) Výsledky paleoetnobotanického studia pěstování zemědělských plodin na území ČSSR. Vědecké práce Československého zemědělského muzea 6:27–144

Tempír Z (1968) Archeologické nálezy zemědělských rostlin a plevelů v Čechách a na Moravě. Vědecké práce Československého zemědělského muzea 8:15–88

Tempír Z (1982) Zemědělské plodiny a plevele z archeologických nálezů v Březně u Loun. Vědecké práce Zemědělského muzea 22:121–195

Tempír Z (1985) Rozbor rostlinných zbytků z Černošic. Archeologické rozhledy 37:14–15

Tempír Z (1992) Analýzy a vyhodnocení zuhelnatělých zbytků kulturních rostlin a plevelů v archeologických nálezech z některých lokalit na jižní Moravě. Vědecké práce Zemědělského muzea 29:73–97

Tempír Z (2012) Rostlinné makrozbytky. In: Kuna M, Němcová A (eds.) Výpověď sídlištního odpadu. Nálezy z pozdní doby bronzové v Roztokách a otázky depoziční analýzy archeologického kontextu: 159–165, Institute of Archaeology in Prague, v.v.i., Praha

205 Trzcińska-Tacik H, Wasylikowa K (1982) History of the synanthropic changes of flora and vegetation of Poland. Zakład Narodowy im. Ossolińskich-Wydawnictwo Polskiej Akademii Nauk van der Maarel E (1988)Vegetation dynamics: patterns in time and space. In: Turner MG (ed.) Temporal and Spatial Patterns of Vegetation Dynamics. 7–19, Springer, the Netherlands

Venclová N (2008a) Archeologie pravěkých Čech 6. Doba halštatská. Archeologický ústav AVČR, Praha

Venclová N (2008b) Archeologie pravěkých Čech 7. Doba laténská. Archeologický ústav AVČR, Praha

Willerding U (1986) Zur Geschichte der Unkräuter Mitteleuropas. Karl Wachholtz Verlag, Neumünster

206